UNIVERSITY OF CALIFORNIA, IRVINE

A Practice Theory of International Law

DISSERTATION

submitted in partial satisfaction of the requirements for the degree of

DOCTOR OF PHILOSOPHY

in Philosophy

by

Daniel David Pilchman

Dissertation Committee: Professor Aaron James, Chair

Professor Margaret Gilbert Professor Richard Arneson

Associate Professor Kevin Olson Assistant Professor Jeffrey Helmreich

2014

© 2014 Daniel David Pilchman

ii    

DEDICATION

To

Claire, Kathy, and David

But not Sara. Definitely not Sara.

As long as you catch self-thrown things it’s all dexterity and venial gain –; only when you’ve suddenly caught that ball which she, one of the eternal players, has tossed towards you, your center, with a throw precisely judged, one of those arches that exist in God’s great bridge–system: only then is catching a proficiency, – not yours, a world’s. And if you then had strength and courage to return the throw, no, more wonderful: forgot strength and courage and had already thrown . . . . . (as the year throws the birds, those migrating bird swarms, which an older to a younger warmth sends catapulting across oceans–) only in that venture would you truly join in. No longer making the throw easy; no longer making it hard. Out of your hands the meteor would launch itself and flame into its spaces . . .

-Rainer Maria Rilke (translated by Edward Snow)

iii    

TABLE OF CONTENTS

Page ACKNOWLEDGMENTS iv CURRICULUM VITAE v ABSTRACT OF THE DISSERTATION vi INTRODUCTION 1 CHAPTER 1: The Concept of Authority in International Legal Practice 15 CHAPTER 2: State Consent 54 CHAPTER 3: Dworkin’s New Philosophy 74 CHAPTER 4: The Practice Approach 99 CHAPTER 5: Conclusion 135 BIBLIOGRAPHY 138

iv    

ACKNOWLEDGMENTS

I would like to express the deepest appreciation to my committee chair, Professor Aaron James, who is as brilliant as he is patient. If philosophers work in the abyss at the edges of human knowledge, he can see in the dark. He taught me always to do the next right thing. Without his guidance and persistent help, this dissertation would not have been possible. Thank you to Margaret Gilbert who has been a source of inspiration to me. She navigates the narrow passage between penetrating genius and modest deference with grace and poise. It has been an honor to work so closely with you. Thank you to Professor Richard Arneson, for taking a confused but enthusiastic graduate student seriously. To Professor Jeffrey Helmreich, for his helpful questions. To Professor Kevin Olson, for teaching me that philosophy can flourish outside of philosophy departments. Thank you to Violet McKeon and Amanda Trefethen. The Justice League has been an invaluable source of support and ideas through this entire process. In addition, thank you to Professor Nicholas Onuf, who is simply an outstanding human being. I thank the University of California for providing my entire undergraduate and graduate experiences, and to the professors and TAs who have taught me all that I know. Thank you also to the Melden Fund, the Matchette Foundation, and the American Philosophical Association for their generous gifts. Finally, thank you to my friends and family. To Claire, mom, dad, Sara, Bryce, and the LBC friends. I love you all.

v      

CURRICULUM VITAE

Daniel David Pilchman

2007 B.A. in Philosophy, University of California, Los Angeles 2007-09 Teaching Assistant, Philosophy, University of California, Santa Cruz 2009 M.A. in Philosophy, University of California, Santa Cruz 2009-2014 Teaching Assistant, Philosophy, University of California, Irvine 2011-2013 Research Assistant to Professor Margaret Gilbert

Department of Philosophy, University of California, Irvine 2013-2014 Adjunct Professor, Philosophy, Chapman University 2014 Ph.D. in Philosophy,

University of California, Irvine 2014 Visiting Assistant Professor, Chapman University

FIELD OF STUDY The Moral, Social, and Political Philosophy of International Relations

PUBLICATIONS “Belief, Acceptance, and What Happens in Groups,” co-authored with Margaret Gilbert, forthcoming in Essays in Collective Epistemology, ed. Jennifer Lackey. Oxford University Press. “Perspectives on the Ethics of Private Asteroid Mining,” forthcoming in Commercial Space Exploration: Ethics, Policy, and Governance, ed. Jai Galliott. Ashgate.

vi    

ABSTRACT OF THE DISSERTATION

A Practice Theory of International Law

By

Daniel David Pilchman

Doctor of Philosophy in Philosophy

University of California, Irvine, 2014

Professor Aaron James, Chair

Why should states follow international law? We are currently witnessing trends

towards rapid expansion of international legal rules and a spreading presumption that

treaties, custom, and other forms of law are binding for states. Despite this, work on the

normative foundations of international law has not kept up. The predominant theories

advanced in the contemporary literature focus on revising law to match some ideal set of

criteria, rather than taking seriously the task of evaluating current regimes. Rather than ask

what law should be, I ask: Is it possible to justify law as it is? In response, I investigate whether

we can construct an independently attractive theory that can justify the authority of

international law as we find it. My inquiry begins with an interpretive analysis of current

practice, to identify both the form and scope of international legal authority. I then offer an

analysis of current theories to see whether any can capture both aspects of current practice.

While they cannot, this process gradually produces a silhouette of what a successful theory

would have to look like. In light of this, I offer a positive answer to the question: the practice

approach to international law. I propose that states have a duty to comply with international

law because of the corrective function that system of rules fills in global politics.

1

INTRODUCTION

At the moment, most practitioners within global politics assume that states are bound by

international law. To some, this may seem obvious in an age when news media front pages tell

stories of how a world leader has flouted international law in a recent state action. While

international law has arguably been around for centuries, if not millennia, the widespread

acceptance of international law as an authoritative source of binding legal norms is something

of a modern development.

At least, it is in the academy. Ronald Dworkin tells us,

When I was last instructed in international law – at Oxford in the 1950s –the first and most lively question, bound to appear on the examination paper together with tedious questions about navigable bays, was existential. Is there any such thing as international law? …the question whether there is international law seems no longer to trouble anyone. Almost everyone assumes that there is international law and also assumes that it includes, for example, the Charter of the United Nations and the Geneva Conventions—or at least some of them.1

Dworkin appears to be satisfied that scholars and practitioners of international law have

[nearly] reached a consensus about the existential question. Yet skepticism persists.

Famously, H.L.A. Hart argued that there is no secondary rule of recognition for primary

international legal rules, thus dismissing the existence of international law.2 More recently,

Jack Goldsmith and Eric Posner have reaffirmed the realist position that international law is

nothing more than the product of states pursuing their own interests, and Thomas Nagel has

argued not only that international law does not exist, - at least not in any normatively binding

way - but also that global justice altogether lacks the kind of justification it would need to bind

international actors in the way that many believe it does.3

1 Dworkin, “A New Philosophy for International Law.” 1 2 Hart, The Concept of Law. 230-231 3 Goldsmith and Posner, The Limits of International Law; Nagel, “The Limits of International Law”; Nagel, “The Problem of Global Justice”. See also Blake, “Distributive Justice, State

2

For Dworkin, the existential question about international law can be interpreted in

two ways: sociologically and doctrinally. The sociological question is a descriptive one; “a

question for social scientists: whether there is any system of practices that can sensibly and

usefully be described, for their sociological or anthropological purposes, as international

law.”4 Few, I imagine, would venture to argue against the existence of international law in the

purely sociological sense. We have documents, drafted and signed by powerful and important

figures and officials, that lay out bodies of imperatives to be followed by states and other

global actors. And in general, or at least within tolerable limits, these commandments are

often respected and followed from a sense of legal obligation. Though I will spend some time

exploring a sociological description of international law, as we find it today, I will mainly

assume that , at least descriptively, international law exists in this sense.

The doctrinal question about the existence of international law is somewhat more

complicated. For Dworkin, international law exists in the doctrinal sense only if it allows

“people to invoke a special kind of right or obligation.” He goes on to say that doctrinal

questions can only be answered with “a theory of political morality about the circumstances

in which something ought or ought not to happen.”5 Questions about the relationship

between doctrinal questions and political morality aside, Dworkin clearly believes that there is

a version of the existential question about international law that is fundamentally normative.

We should understand the difference and relation between the sociological and

doctrinal questions about the existence of international law in the following way: The

sociological question can be answered by reference to empirical evidence that agents act with

certain attitudes, as though they are legally obligated to comply with certain bodies of Coercion, and Autonomy”; Cohen and Sabel, “Global Democracy”; James, “Distributive Justice without Sovereign Rule.” 4 Dworkin, “A New Philosophy for International Law,” 4. 5 Ibid., 11.

3

imperatives, presented in socially salient documents. The doctrinal question can only be

answered with a normative theory that justifies that putative obligation – a theory that argues

for the claim that international agents are the genuine bearers of the obligations that they act,

sociologically, as though they have. Put otherwise, and in the language that I will use

throughout this essay, the sociological question asks whether there are norms that

participants in international political life treat as legally authoritative, and the doctrinal

question is whether and how norms could genuinely or justifiably be authoritative in that

sense.

By framing the question this way, especially the sociological question, I make a strong

presumption in favor of the existence of international law in the sociological sense. This

presumption marks a significant methodological departure from modern legal scholarship by

according a privileged position to the practice and self-understanding of practitioners. Some,

likely including Hart and perhaps Fuller, will object to the existence of international law –

Hart because he thinks there is no obvious secondary rule of recognition; Fuller because his

specified structural and procedural standards are not met.6 But, as Hart happily concedes,

there is clearly a body of rules that regulates international political life, that is followed from a

sense of legal obligation, that is used by international courts and tribunals, and that exhibits

other features shared with domestic law.7 Rather than take the differences (“disanalogies,”

Hart might have called them) between domestic and international law to be evidence that

international law is not actually law, I propose we use these differences to rethink what it is to

be law. In a phrase, Hart's modus ponens that international law is not law is my modus ponens

that Hart’s concept of law should be broadened.

6 Hart, The Concept of Law, Ch 11; Fuller, “Eight Ways to Fail to Make Law,” 33–38. 7 Hart, The Concept of Law, 228–229.

4

Given this presumption, my project takes an interest in both the sociological and

doctrinal questions about international law, specifically as it currently exists. I am interested

in the doctrinal question as it applies to the modern context, which in turn means that I will

have something to say about the state of international law in that context. More and more,

scholars and practitioners agree that states have an obligation to comply with international

law, that the law has authority over them, that it is wrong to disobey or at least that non-

compliance can be justifiable grounds for censure, and so on. My question is how this could

be. How it could be that states have such an obligation? How could international law, as we

find it, be authoritative?

My answer will come in four chapters, each taking successive steps towards a positive

analysis. In Chapter 1, I interrogate the concept of legal authority in two venues - first as it is

discussed in philosophical literature, second as it is understood by legal practitioners. In the

literature, authority, speaking loosely, is what makes laws reason-giving for agents. If the

traffic laws in my neighborhood are authoritative, then I ought to obey the rules of the road

under normal circumstances. Authors, with important exceptions, frequently take this

concept for granted - something that competent users of language understand well enough to

support meaningful discourse. However, a sustained interrogation of the varied uses of the

concept of authority across different legal arenas - and specifically ones that propose to

answer the doctrinal question - suggests significant semantic diversity. This diversity seems

not to be accidental, but connected to the other philosophical commitments implicit in these

competing views.

Given the apparent intractability of disagreement between competing doctrinal

theories of law – especially in the long stand-off between natural law theorists and legal

positivists – the recognition of diversity in usages of the concept authority opens a promising

5

line of inquiry. If we had independent reason to prefer one concept of authority over the

other, then this preference could serve as the basis for an argument for one answer to the

doctrinal question over another. More precisely, any theory that could not justify legal

authority in our preferred sense would be at a philosophical deficit, and this might be enough

to make some progress on old questions.

But is there any independent reason to prefer one concept of authority over another? I

believe there is. Philosophers, including legal philosophers, are given a fair amount of latitude

in how far their imaginations are allowed to run. A person could spend an entire career

writing about the concept of authority in an unlikely utopia and her colleagues would not bat

an eye. But there is, I believe, independent reason to be interested in the concept of authority

presumed in modern practice. This is because, while there are many worthy reasons to engage in

political philosophy, one of the most best and most important is to assist and promote good

practical reasoning, and this is only accomplished when we engage with the world reasoning

agents find it.

Questions about the authority of law are nothing new to political philosophy or to

political life, and there is a long tradition of authors who wrestle with them. It hardly needs

saying that some of the most illuminating and cherished philosophical work about the

authority of law and the corresponding obligation to obey (or whether there is one) come to us

from Socrates himself. But this thread of argument marks a methodological shift for legal

philosophy. For the most part, the concept of authority has been a kind of byproduct of

philosophizing about law. I propose to reverse that ordering and allow arguments about the

appropriate concept of authority to constrain whatever else we say about law.

To that end, the latter half of chapter 1 is an interpretive analysis of current

international legal practice, with special attention paid to the kind of authority that

6

international law is generally presumed to have. More specifically, I argue that the concept of

legal authority implicit in modern practice has three basic characteristics: when law is

authoritative, it (1) gives states reasons for action, (2) which are defeasible in overall practical

reasoning, and yet (3) binds states independent of the moral (or immoral) content of any given

legal rule. Summarized more succinctly, when international law is authoritative, agents ought

to comply, whatever it commands, unless they have particularly good reason not to. If this is

indeed the kind of authority that law is presumed to have in international legal practice, the

question, then, is: how could international law, as we find it, have that?

One important result from this line of inquiry is the identification of a particular

objection to natural law theories. By denying the separability thesis - that legal rules may

contradict moral rules - natural law theories deny that states might have legal obligations to

comply with immoral laws. This means that, at least without further explanation, natural law

theories seem unable to capture the content-independent character of legal authority in

practice. I will explore one way that natural law theorists have tried to overcome this obstacle

in Ch. 3.

Chapter 2 is a first pass at an answer to my question based primarily on the familiar

(and almost universally rejected, at least in the academic literature) state consent thesis: that

international law is authoritative only if and because states consent to it. It turns out that this

view, with deep roots in the legal positivist tradition, can be augmented to incorporate the

kind of authority implicit in practice. Given my favored methodology, this speaks in favor of

taking state consent seriously as a contender to answer Dworkin’s doctrinal question. Indeed,

given the widespread popularity of social contract views in political philosophy about

domestic issues, it is surprising that the state consent thesis - a close cousin - would be so

readily discarded, as it seems to have been.

7

Despite my optimism, sustained discussion of state consent theory as an answer to the

doctrinal question about international law shows that it does fail. However, that it fails is of

less interest to my project than how it fails. State consent theory is indeed a way to justify a

legal system that wields the form of authority we are looking for, so, in that regard, counts as

successful. The problem with state consent theory is not one of form, but rather one of scope.

As so many authors have shown, not all of international law can be rightly described as the

object of state consent. So accepting state consent as an answer to the doctrinal question

forces us to be extremely revisionistic about the sociological question. Perhaps this is not so

bad. Surely, one might think, the international legal system is an imperfect conglomeration of

bits and pieces of custom and precedent and, like any legal system, it could bear a bit of

pruning. But, I will suggest, this would vastly understate the extent to which state consent

theory would demand revision of present legal rules - even those widely accepted as law and

as legally authoritative.

Speculating historically, this shortfall in scope may be due to the close relationship

between state consent theory and social contract theory. Most saliently, social contract theory

- by far the more fully developed of the two - is concerned with domestic society. This

orientation towards the domestic, specifically apparent in the examples and analogies taken

to be persuasive evidence for state consent theory and for legal positivism more generally,

obscures the fairly large differences in the kinds of rules authorized by the theory and those

presumptively authorized in practice. This inherited parochialism accounts for why and how

the scopes of state consent theory and of real practice might have diverged from one another.

Of course, the fact that state consent theory is revisionistic is not itself a knock down

argument against the view. It may be the case that international law as we find it cannot

justifiably bear the kind of authority it is presumed to have, but this would have to be shown.

8

So just as there is independent reason to wonder about the concept of authority we find in

practice, the same reason supports the search for an alternative that defends the concept of

authority in practice and is minimally revisionistic. So the question we started with persists,

though now with further specification. If at all, how could international law as we find it have

the kind of authority it is presumed to have?

Chapter 3 is a close analysis of Ronald Dworkin’s posthumously published “New

Philosophy of International Law”, which is among the most promising answers to precisely

this question. We find an important ambiguity at the core of this view, and one interpretation

of the new philosophy (though likely not Dworkin’s favored interpretation) does answer the

doctrinal question of the authority of international law.

Dworkin argues that international law is authorized by the corrective function it plays

within international political life. He argues, contrary to the state consent theorists but at least

consistently with prominent constructivist authors in international relations theory, that

states owe their sovereign rights to their participation in the state system, and the justifiability

of their claims to those rights to the legitimacy of that system. The catch is that the state

system is prone to a number of morally salient and negative side-effects of distributing

political power the way that the state system does. For example, by organizing international

political life such that states are given wide-ranging rights to domestic legislation and

enforcement, the international community leaves minority groups especially vulnerable to

oppression. Similarly, by granting states rights to extract and use resources at will, the

international community creates coordination problems with respect to cross-border

problems like atmospheric degradation and pollution.

These pathologies undermine the legitimacy of the state system, but can be corrected

for through various institutional instruments including, especially, international law. The

9

oppression of minorities can be stopped when states comply with, and mutually enforce, legal

standards of conduct like the UDHR. Climate change and environmental degradation can be

addressed by states negotiating, complying with, and enforcing treaties like the Kyoto

Protocol (which is a particularly interesting example). By complying and building assurance

of future compliance, states can address pathologies and legitimate the state system.

But how does this translate into a state’s obligation to comply with international law -

what Dworkin calls the “duty of mitigation” - rather than just something that it would be nice

or even advisable for the state to do? His answer to this question is, I believe, where the

ambiguity arises. It might be, as Dworkin sometimes seems to imply, that states have a duty to

improve their own legitimacy because citizens have a correlative natural right to legitimate

governance. This would be consistent with Dworkin’s established commitment to his own

version of natural law theory as well as his skepticism about positivist views. On this view,

states would have an obligation to improve their own legitimacy, which can (only?) be done by

improving the legitimacy of the state system. This, in turn, can only be done by complying

with the most “salient” solution to the pathologies of the state system, which, Dworkin

believes, is international law.

Matthias Kumm offers another interpretation of the new philosophy. While the above

version depends on the state obligation to improve its own legitimacy, Kumm grounds his

version in the state obligation of non-domination. When states take the resolution of

pathologies of the state system into their own hands, without consulting other states through

proper procedural channels, they are essentially dictating the issues that affect many states

will be resolved. This, Kumm believes, is a morally objectionable form of domination. Only be

addressing pathologies through international law, supposing procedural safeguards that

make sure the international legislative procedure is not itself a form of domination, can

10

pathologies be permissibly addressed and the legitimacy of the state system upheld. Thus law

is authoritative for states not (necessarily) because they have obligations to their citizens, but

because they have obligations of non-domination towards one another.

Both of these interpretations offer natural law theories, though with some finessing of

the separability thesis. Dworkin is willing to allow that what is presumed to be international

law is law in the sociological sense, but has reservations about its authority, in the doctrinal

sense. That is, the norms states recognize and treat as law are indeed law, but they are not

authoritative for states and state actors unless compliance with them helps states satisfy their

natural duty to improve their legitimacy. By ultimately linking the authority of law to the

natural duties of states, Dworkin (and Kumm) run into what I have called the problem of

form: legal duties are ultimately moral duties, and so the possible content of authoritative law

depends on its relation to true moral claims. But, again, this is not the content-independent

form of authority we find assumed in practice, in which case we can take objection to

Dworkin’s view.

Chapter 4 offers another way to interpret the ambiguity in Dworkin’s stated position,

although it may be more correct to say that Dworkin’s attention to the state system’s

pathologies points us in a new and fruitful direction towards what I call the practice theory of

international law. According to this view, established international law is authoritative for

states (and states have an obligation of compliance) insofar as it corrects for the pathologies of

the state system. Unlike other salient problems, the state system’s pathologies are caused by

the state behavior that also generally constitutes the state system. The fact that agents are

involved in a pathological but largely cooperative social practice suffices to justify an

associative obligation of mitigation, which obliges states to comply with international law

much as Dworkin envisions.

11

To briefly elaborate, the following is a sketch of my version of the new philosophy as

grounded in associative obligations. In our world, sovereignty is an institutionally defined

right to rule distributed amongst a decentralized group of territorially bounded protective

associations. These associations (i.e. pre-states, or nominal states, or state-like-entities-prior-

to-recognition) gain the sovereign right to rule only by being recognized as a state by other

states. Together, these states populate the state system, a social practice that both distributes

and legislates the terms of rule. This system is presumed to be good for many reasons,

including the fact that it promotes shared and widely acceptable goals, such as prosperity,

security, and toleration. But by pursuing these goals through such a decentralized system of

governance, the system also has harmful side effects.

Two basic means have emerged within the state system to address these pathologies,

or untoward tendencies. First, the state system has developed a number of mechanisms for

creating norms general compliance with which would help mitigate these pathologies. Among

these mechanisms are the sources of law specified in Article 38 of the UN Charter - treaties,

custom, and so on. Second, the state system has - through these mechanisms - developed a

number of regulatory norms presumed to bind states such that, absent special countervailing

considerations, they ought to comply with those norms regardless of what they command.

When states claim the right to rule - the rights of sovereignty institutionally specified by the

state system - they become members of this problematic but presumptively justifiable

practice, and as members, undertake an obligation to the other members to do their part in

mitigating the system’s problems. In the current world context, that obligation can only be

discharged by complying with international law.

A practice theory of this kind not only offers a logically consistent answer to the

doctrinal question of law, it captures both the form and scope of law presumed in practice, at

12

least better than the considered alternatives. First, it explains how law can give the kind of

reasons it is presumed to give both from an “external” and an “internal” perspective. From the

outside, the practice approach offers an argument that states have associative obligations to

other states to comply, which, like any obligation, ought to suffice as a reason for action. From

the internal perspective of states themselves, the practice approach links whatever reasons

they might have to claim the rights of sovereignty — power, wealth, clout, status, etc — and

links those reasons to the need to uphold and improve the legitimacy of the state system, since

that system is the source of those rights. Compliance with international law is the only means

available to states to ensure their claim to the rights of sovereignty. This account of the

reason-giving character of law also allows us to distinguish legal imperatives from advice and

requests, since heeding advice and granting requests is not “owed to” others the way that

associative obligations are.

While law’s grounding in associative obligations may ground a kind of reason for

action, other more familiar cases — like familial duties — show us that associative obligations

can sometimes be overridden or defeated by sufficiently grave countervailing concerns. Thus,

while law gives reasons for action, it may yet be defeasible by other countervailing norms.

Finally, legal authority, on this view, is content-independent in the sense identifiable

in practice. In the first place, the state obligation to comply does not depend on whether the

content of a legal rule corresponds with widely accepted moral truths. In fact, it is even

conceivable that certain legal rules might be immoral in isolation, but nevertheless play an

important corrective function in the overall global political practice. Thus, for example, a

bilateral agreement to liberalize trade might give wealthier countries certain economic

advantages, which is morally objectionable from an egalitarian perspective, but be part of a

larger international legal movement towards global trade liberalization that pursues the

13

purpose of global prosperity. So even when the law asks them to do something immoral, if the

law fulfills its corrective function in the overall picture, then that law is authoritative for states

and they have an obligation to comply.

The practice view is also better able to capture the scope of international law as we

find it, even if it may not capture all of the rules treated as authoritative. It explains legal

authority and illuminates the interpretation of the paradigmatic cases raised in Chapter 1,

cases that raise problems for competing theories. In this regard, the practice view ought to be

preferred on the basis of being minimally revisionistic for these cases. But I also argue that

only the practice view can accommodate rules treated as law from every one of the recognized

sources of law. Positivist views have difficulty explaining customary international law and

much of administrative international law, including jus cogens norms, human rights, and the

pacta sunt servanda principle. These blind spots force positivists to render vast parts of

international law lacking in authority. Natural law views struggle to explain treaty and

administrative law without supposing a fairly ad hoc set of “natural” states rights and duties.

The practice approach avoids these problems by linking authority to the function legal

rules play in a larger social practice rather than the source from which the rule arises.

Positivist views are highly revisionist because they only authorize laws that come from

sources that are sensitive to consent, while natural law views are revisionist because they only

authorize laws that come from sources sensitive to morality. The practice view is, in this

respect, more pragmatic. It matters less where a law comes from and more whether it helps

legitimize the legal practice overall. Law is, above all else, a tool. And just like any other tool, it

is good or bad, better or worse, in virtue of its ability to do what we need to use it for. In the

case of international law, that function is to govern global political life so that the state system

14

might be freed from charges of illegitimacy, for forcing states to coexist on terms that are not

mutually acceptable.

15

CHAPTER 1: The Concept of Authority in International Legal Practice

Introduction

Why should countries comply with international law? One simple answer, bordering

on tautology, is that states ought to conform international law because it is a legal system that

has authority over them. Without further analysis, this answer would neither illuminate our

question nor seem promising as an answer to dissenting and obdurate participants in global

politics. The bald assertion that international law has authority passes the buck, as it were. It

simply raises the question: But why does international law have authority for states? The

assumption, both common and easy to make, is that these questions are nearly identical, and

that the latter cannot illuminate the former.

This assumption is a mistake because an analysis of the authority of international law

can in fact illuminate and offer guidance to an investigation of why states ought to comply

with international law. How? Because different answers to Dworkin’s doctrinal question will

either imply or presuppose different accounts of legal authority. These implications are not

merely judgments that positivists, natural law theorists, and others arbitrarily make as a

supplemental consideration. They are necessary consequences of the other commitments

these theorists make in describing their views, and they can be evaluated for their plausibility

and fit on their merits.

The recognition of the intimate relationship between competing legal theories and

their implicit concepts of authority raises the possibility of using a favored understanding of

authority as a basis for adjudicating between competing views. If there were some conception

of authority for which we had independent reason to judge appropriate to a theory of

international law, then how well a theory is able to capture that sense of authority would serve

16

as a basis for adjudication. More simply, a theory of international law that could not capture

the preferred kind of legal authority has a problem, or at least is at a comparative deficit

relative to a theory that can.

More than allowing adjudication between theories, such a preferred conception of

authority could guide the development of new theories. By focusing philosophical work on

that particular kind of authority and trying to understand its basic characteristics, we could

refocus normative work on thinking about how best to justify a body of norms with those

characteristics. This would offer an entirely new set of handholds in a notoriously slippery

climb.

All of this of course depends on our ability to identify a conception of authority that we

have independent reason to prefer. But do we have any such conception? A skeptic might

remind us that different projects in legal theory may well need to make use of different

conceptions of authority. Sure, the values that underlie their respective goals could serve as

the kind of independent reason to prefer one concept to another for that project. The concept

of authority most appropriate to a theory of law in Nazi Germany, for example, will almost

certainly be distinct from that most appropriate for a theory of utopian law. But, the skeptic

might continue, because different projects have different purposes, no one conception of

authority stands out above the rest. So long as theorists are free to choose their projects, and

different projects are equally valuable, there is no independent reason to prefer one

conception of authority to another.

At the same time, unlike many areas in philosophy, political and legal theory are

largely motivated by real practical problems like human rights abuse, poverty, hunger, the

threat of nuclear war, and the belief that we can contribute to the resolution of those

problems by offering resonant analyses to guide practitioners. Contributing to the resolution

17

of these problems is arguably a more important task than philosophical abstraction or

historical speculation. Therefore, if there is a project in legal theory that has this purpose,

then this purpose should serve as the independent reason to prefer the concept of authority

implicit in that project. The question, then, is What might such a project be?

The project I propose is a philosophical investigation into why states ought to comply

with international law as we find it. Today, in our world, international law represents one of

the best chances we have for making real progress on precisely the practical problems

mentioned earlier. And in any case our understanding of why countries have an obligation to

comply with international law as we find it is at best underdeveloped. In fact, states do

generally comply with international law, or at least their actions are observably shaped by

their presumption that it has a kind of authority over them. This assumed conception of

authority can be studied. Once properly interpreted, the question is then, Can we justify that

kind of authority in the modern context?

I believe we can. This chapter begins the investigation into the character of

international law as we find it and the conception of authority implicit therein. I begin with an

analysis of international law itself, specifically focusing on what are commonly referred to as

the “sources” of international law: treaties, custom, court rulings, and administrative

commands. Before offering an interpretive argument about the characteristics that authority

is presumed to have within this legal system, I turn to existing theories of law and legal

authority to get some sense of how others have characterized the authority of law in the past.

This review illuminates the kinds of characteristics typical of a theory of authority, and so

gives some sense of the kinds of questions we should ask during an interpretive analysis of the

modern international legal system. This will also foreshadow extended discussions of the

predominant theories of international law that follow in chapters 2 and 3.

18

Finally, I offer my interpretive analysis of the concept of authority implicit in modern

international law. Central cases, I argue, reveal three main characteristics of authority. First,

international law gives states directed reasons for action that are distinguishable from the

kinds of reasons given by advice and requests. State behavior is observably shaped by

awareness of the legality or illegality of an action, and a state’s failure to comply is presumed

to be appropriate grounds for accountability seeking behavior by other states. This indicates

that practitioners presume (a) that states are the primary — though perhaps not the sole —

subjects of international law, (b) that states owe conformity to other states in some sense, and

(c) that non-compliance is a kind of wrong, though not necessarily a moral wrong.

Second, while international law is presumed to give states reasons for action, these

reasons can be overridden by sufficiently weighty countervailing concerns. For instance,

many countries fail to comply with human rights standards, but we can observe variations in

the kinds of accountability seeking behaviors that other states react with. These variations

appear to track the material capabilities of non-compliant states such that human rights

violations in developed countries are treated more harshly than in developing countries. The

presumption appears to be that although states ought to comply with the law, that duty or

obligation can be defeated by other considerations.

Finally, states are presumed to be bound by international legal rules even when they

are inconsistent with widely accepted moral truths. Legal authority is, in other words,

“content independent.” The bindingness of international law does not depend on the

morality or immorality of what it commands. The North American Free Trade Agreement

(NAFTA) is, for example, an apt target of significant and, I think, justified moral criticism.

Nevertheless, these legal rules are recognized as law and, as such, hold a kind of authority

over states.

19

These three characteristics — directed reason giving, defeasibility, and content

independence — form the core of my conception of the legal authority implicitly presumed in

international practice. Collectively, they establish standards for adjudicating between

competing theories, and offer guidelines for the development of new alternatives, a task that

will occupy the subsequent chapters of this project.

The Scope of International Law As We Find It

When legal scholars and practitioners use the term “international law”, what do they mean?

We might answer this question in either or both of two ways. First, we might try to list the

objects to which people refer when they use the term – that is, we might offer the extension or

scope of international law. Second, we might offer a conceptual analysis of the concept that

those speakers invoke – a list of the general characteristics of a kind of norm. Let us begin with

a discussion of the former – the scope of international law – in hopes that it will lead us to the

latter.

It has become common practice among legal scholars, when discussing the scope of

international law, to abbreviate what would be a long list of principles, customs, and statutes,

to a shorter list of the sources of international law. The assumption is that if we can identify

the sources of law, then the scope of law is just whatever has been or will be produced by

those sources. Nowhere are the sources of international law, at least in the modern context,

more clearly or definitively spelled out than Article 38.1 of the Statute of the International

Court of Justice within the UN Charter, which reads:

The Court, whose function is to decide in accordance with international law such disputes as are submitted to it, shall apply: a. international conventions, whether general or particular, establishing rules

expressly recognized by the contesting states; b. international custom, as evidence of a general practice accepted as law; c. the general principles of law recognized by civilized nations;

20

d. subject to the provisions of Article 59 , judicial decisions and the teachings of the most highly qualified publicists of the various nations, as subsidiary means for the determination of rules of law.8

This statute identifies three unique sources of international law: (1) conventions and treaties,

(2) general customs and recognized principles, and (3) court rulings.9

Treaties

Treaties are bi- or multi-lateral agreements between states. Parties may and do negotiate the

terms of the agreement, until a mutually acceptable version of the agreement is discovered.

When both or all parties give the appropriate signs of consent (for the US, this includes

legislative ratification domestically), treaties are subsequently recognized as sources of legal

rights and obligations for signatories.

International legal scholars frequently treat treaties as the paradigmatic example of

international law, and this preeminence can be speculatively attributed to the concordance

between the doctrines of state consent at the international level and popular sovereignty at

the domestic level.10 Domestically, Hobbes, Locke, and others teach us that subjects come to

have legal obligations by consensually conceding some measure of their natural rights to the

8 United Nations, “Article 38.1 Statute of the International Court of Justice.” 9 I make two interpretive claims here. First, I collapse general customary practice and general principles, because I take these to be mutually constitutive if not simply identical. Second, I collapse the “teachings of highly qualified publicists” into court rulings. Scholars, it seems to me, can offer practitioners of the law novel ways to interpret the law, and this interpretive function is normally attributed to the courts. However, as will be come apparent in subsequent sections, it is not obvious to me that such interpretation only happens in judicial fora. In fact, it seems very likely that all state action, including the treaties and customary practice, is an interpretation of international law and the boundaries of permissible activity. Perhaps, then, it is better to say that the relevance of scholarly work as a source of law is implicit in all sources of law, but is not obvious a source independent from the others. 10 Goldsmith and Posner, The Limits of International Law, 23–24; Brierly, The Basis of Obligation in International Law, 9, 11, 79.

21

society for the sake of security, collective welfare, etc.11 On such a view, the basis for legal

obligation is the consent of the governed. Treaties allow us to tell a similar story about

international law: they are consensual contracts that limit state rights, but only so far as those

states will.

The Convention on the Prohibition of the Development, Production, Stockpiling and

Use of Chemical Weapons and on their Destruction (hereafter Convention on Chemical

Weapons or CCW) is a helpful example of a modern treaty. Originally adopted in 1993, the

CCW represents an international effort to “to eliminate an entire category of weapons of mass

destruction”.12 To date, 190 countries around the world have ratified this agreement, notably

including Syria in 2013.13 Signatories are obligated to other parties to the agreement to

promote the goal of elimination and security, and are vulnerable to legal sanction in case of

noncompliance.

Custom

The second source of international law is custom or customary practice. Customary law can

be generally described as the norms and principles that are not necessarily codified in a treaty

of convention, but are instantiated by regular state behavior done from a sense of legal

11 Hobbes, Leviathan; Locke, Second Treatise of Government. 12 Organization for the Prohibition of Chemical Weapons, “Chemical Weapons Convention.” 13 The Syrian case is of particular relevance for two reasons. First, while the Syrian government was widely criticized for its use of chemical weapons against its own citizens in early 2013, its actions were not illegal, at least not with respect to the CCW. This fact inhibited the political and legal authority of observers inclined to intervene. Intervention in the form of the coerced removal and destruction of Syria’s chemical weapons stockpiles (currently underway) only began after Syria ratified the CCW in September, 2013. Second, Syria ratified the CCW only under extreme pressure from the international community. Thus the Syrian case serves as an example of a treaty that maintains legal authority despite the fact that consent was coerced.

22

obligation.14 Some familiar examples of customary international law, or at least salient

examples from legal history, are the extension of territorial boundaries into international

waters (which expanded alongside the reach of cannons), and the special legal immunities

afforded foreign diplomats.

Unlike treaties, customary international law need not be the product of formal

legislative procedures. The laws constitutive of treaties are generally highly specific, explicitly

negotiated, and consciously drafted. Customary law need not, and frequently is not, like this.

Legal customs typically form gradually. They change and emerge with changing and

emerging technologies, needs, and cultural norms.15

Some customs, whether because they are of particular importance or controversy,

have been codified in the form of conventions. For example, the modern standards for

diplomatic immunity, though a longstanding and widely recognized tradition, were codified

in the 1961 Vienna Convention on Diplomatic Relations.16 Something similar can be said

about the state’s right to make treaties which, though a long standing tradition, was codified in

its current form as late as 1986.17 Other customs are so basic to international relations and

social life that they do not warrant codification. The customary principle pacta sunt servanda,

for example, is so fundamental so social life, and so widely followed and recognized, that

14 Cassese, International Law; Goldsmith and Posner, The Limits of International Law. The language of acting “from a sense of legal obligation” is interesting for at least two reasons. First, in the context of international law, we are normally talking about corporate actors, groups. It would be unusual to attribute sentiments to groups, since they are generally not through to have minds, and therefore the capacity to have feelings. Second, although this sense of legal obligation is usually only discussed with respect to customary law, one might expect that international agents act from such a sense when they comply with treaties and court rulings as well. 15 This need not always be the case, however. Bin Cheng provides an interesting example of “instant custom” regarding territorial border in airspace during the cold war. See Cheng, “United Nations Resolutions on Outer Space.” 16 Denza, “Vienna Convention on Diplomatic Relations.” 17 Zemanek, “Vienna Convention on the Law of Treaties.”

23

scholars and practitioners allow it to be taken for granted as a general expectation in

international relations.18

Court Rulings

As in the case of domestic law in most countries, judicial decisions are a source of

international law. It is the widely accepted function of judges and tribunals to interpret the

outputs of these sources of law (treaties, customs, previous rulings), and to determine how

those ought to be applied to novel cases brought before them. In the modern context, the

preeminent international courts are most likely the International Criminal Court (ICC) and

the International Court of Justice (ICJ).

One significant complication in trying to establish court rulings as a distinctive source

of international law is that these predominant examples, the ICC and ICJ, are both the

products of treaties – the Rome Statute and the UN Charter respectively. This fact makes the

legal authority of court rulings appear to be dependent on the legal authority of treaties. But

there is, I think, better reason to reject such a claim. Consider the ICC, probably the most

widely recognized international judicial body with the jurisdiction to prosecute individuals

(the ICJ does not have this legal right). If rulings of the ICC only bound signatories to the

Rome Statute, then the ICC could not possibly make authoritative rulings over individuals.

But the prosecution of individuals is precisely the function of the ICC, and one that is widely

accepted as justified. So at least speaking sociologically, it is, I think, helpful to distinguish

international courts as a distinct source of international law.

18 Thinking of pacta sunt servanda as a legal custom makes it seem strange that, in modern scholarship, treaties should be seen a paradigmatic and customs problematic. As Lauterpacht points out, treaties are entirely dependent on custom, a point to which I return. See Lauterpacht, The Function of Law in the International Community, 420–422.

24

Administrative Law

An additional source of international law, not explicitly recognized in the UN Charter,

is administrative international law (AIL).19 The fact that this source is not recognized in the

UN Charter is not necessarily an indication that AIL is not a kind of law, but a reminder that

the UN Charter is not the secondary rule of recognition for international law.

Administrative law is a growing body of rules for the regulation of global political life,

designed by specialized non-government organizations, and followed from a sense of legal

obligation. One prominent example of administrative law is the International Labor

Organization’s labor standards. These standards outline the appropriate treatment and

working conditions for employees in various industries around the world. Failure to meet

these standards is widely perceived as a source of national shame and a wrong.

Administrative law is unique in that it has many of the formal and procedural

elements of treaties – it comes about from the deliberate and conscious efforts of individuals

and groups, for instance – but can emerge without the consent of the states that are

subsequently bound. It might seem strange to say this, since even in the case of the ILO, states

are only formally bound to the labor standards after domestic ratification. But by garnering

widespread support, invoking widely accepted values, and raising awareness of improper

practices, standards like the ILO’s can come to exert the same sort of reason-giving force that

treaties and other forms of international law do, until it seems more clearly just to be one

more source alongside the others. That said, administrative law is certainly an emerging (if

not fully developed) source of international law, so it may yet be inappropriate to call it a full-

fledged source.

19 Kingsbury, Krisch, and Stewart, The Emergence of Global Administrative Law.

25

What is legal authority?

The preceding was a discussion of the content or scope of international law as we find it. A

complete description of international law will also describe its form. International law does

not present itself simply as a set of inert norms. It is a set of commandments that seems to

demand action or at least purports to be action-guiding in some sense. International law

presumes (and, I will later argue is presumed to have) its own legal authority. But what is legal

authority? Using broad strokes, Joseph Raz induces a sense of perplexity when he writes,

Sometimes things are referred to as authoritative simply to indicate that they are binding, perhaps because they can be trusted to guide one correctly, they are dependable. ‘The authority of reason’ is perhaps used in this way… This, however, is an extended and watered down use of ‘authoritative’. In its central meaning the authoritative is what was made binding or reliable by an authority. An authoritative edition of a poet’s work, or the authoritative text of his poems, or the authoritative version of it, are those which were prepared by authorities in his work…20

Raz employs at least three different senses of authority. First, in the phrase “the authority of

reason”, authority is a quality of a form of practical deliberation (reasoning) such that is

prescriptions are appropriately treated as reasons for action. Second, a text can be

“authoritative” in that it is best version, or the most widely recognized, or the one preferred by

the creator. Third, an agent can be authoritative or an authority when he or she has the

standing to make demands on or otherwise change the normative circumstances of others.

Thus a literary author has authority over his or her work – that is, the normative power to tell

others which is the correct interpretation. And a police officer has authority over individual

comportment in public spaces- the normative power to command and even coerce legally

compliant behavior like driving the speed limit.

But when we speak of the authority of law, it is not obvious that any of the senses listed

above captures precisely what we mean. As in the first sense, authority is a quality associated 20 Raz, “Introduction,” 1–2. Compare Hart, The Concept of Law, 198–199.

26

with practical deliberation. But unlike reason, law is a body of norms, not a form of

deliberation. As in the third sense, the fact that law is authoritative seems to give it the

standing to make demands of its subjects. But again, that is not quite right. Law is not an

agent, unlike the police officer and the literary author. So while these distinctions are helpful

in getting a feel for the concept for which we are casting about, they offer little towards really

understanding what legal authority is.

One important distinction that we should establish right at the outset is that between

genuine and false authority. This distinction is a familiar one. Genuine authority is a

command from a norm or an agent that actually gives one a reason for action. False authority

fails to give reasons for action, but acts as if it does. So while a police officer may have the

authority to give me a speeding ticket, a person merely dressed as a police officer, on his way

to a costume party, who gives me what he calls a “citation” does not have the genuine

authority to do so. Raz captures this distinction perfectly, though in different terms, when he

writes:

Authority in general can be divided into legitimate and de facto authority. The latter either claims to be legitimate or is believed to be so, and is effective in imposing its will on many over whom it claims authority, perhaps because its claim to legitimacy is recognized by many of its subjects. But it does not necessarily possess legitimacy.21

I will reserve the term ‘legitimacy’ for something more specific later in this dissertation, but

Raz’s idea is clear enough. Many agents and norms act as though they have authority, but only

some really do. Hereafter, I will use the term ‘authority’ to stand for genuine authority

(legitimate authority, in Raz’s terms), and will specify if I mean otherwise.

Natural Law: Legal Authority is Moral Authority

21 Raz, “Authority, Law, and Morality,” 296.

27

A review of the extant literature on legal authority reveals a variety of accounts, most of which

tend towards either of two poles. At one end, authors identify legal authority with moral

authority and say that law exerts the same kind of normative force as correct moral principles.

On the other end, theorists claim that legal authority has a specific character, distinct from the

kind of authority that true moral principles wield.

Views in the first camp – those who claim that legal authority is a version of moral

authority – are primarily associated with the natural law tradition. On the relationship

between morality and law, John Finnis, a distinguished proponent and defender of natural

law theory, writes

If we may translate the relevant portion of, for example, Thomas Aquinas’s theory into Kelsenian terminology (as far as possible), it runs as follows: The legal validity (in the focal, moral sense of ‘legal validity’) of positive law is derived from its rational connection with (i.e. derivation from) natural law, and this connection holds good, normally, if and only if (i) the law originates in a way which is legally valid (in the specially restricted, purely legal sense of ‘legal validity’) and (ii) the law is not materially unjust either in its content or in the relevant circumstances of its positing.22

Elsewhere, specifically writing about authority in general, Finnis writes:

…in some forms of human community, that something be done [about coordination problems] is not just a matter of optional advantage, but is a matter of right, a requirement of justice… There are, in the final analysis, only two ways of making a choice between alternative ways of co-ordinating action to the common purpose or common good of any group. There must be either unanimity, or authority.23

Here, Finnis seems to claim that whether or not something has authority is a matter of

political morality. It is morally imperative, “a matter of right”, that humans establish and

adhere to systems of rules to coordinate their behavior on salient issues. This duty is

discharged by establishing an authoritative regime to regulate coordination, and is the basis

for the authority of that regime and its commandments. Elsewhere, Finnis explicitly extends 22 Finnis, Natural Law and Natural Rights, 27; Aquinas, Summa Theologica, Part I-II, Q. 96, Art. 4. 23 Finnis, “Authority,” 175.

28

this distinctly moral conception of authority to law. Discussing the conception of authority

contained in Dworkin’s Law’s Empire, Finnis sympathetically writes about “the legal and thus

the moral authority of most of the law's rules and institutions,” which is perhaps an even

clearer statement of the idea that legal authority appears, for Finnis, to be a kind of moral

authority.24

What would it mean for legal authority to be a kind of moral authority? It has been

alleged, primarily by those opposed to natural law theory, that such a thesis would imply that

human law is nothing more than a copy, an “emanation”, of natural law.25 Kelsen is prone to

such antagonistic ascriptions, and the associated skepticism about whether all of law must or

even could be derivable from a full account of the moral truths of the universe. He writes:

An unbiased analysis of the natural-law doctrine shows that it is impossible to deduce from ‘nature’ any rights. For the right of an individual presupposes the duty of another individual, and nature, that is, a complex of facts determined by the laws of causality, does not impose duties and therefore does not confer rights upon men or other beings.26

Finnis, in reply, calls this “a travesty”. But even without going as far as Kelsen, we might say at

least the following: If legal authority is a kind of moral authority, then immoral laws are

neither legally nor morally authoritative. Assuming that morality is internally consistent with

itself – which follows from the idea that ought implies can – it could not be the case that a

legally (and so morally) authoritative norm was inconsistent with some other morally

authoritative norm. It would be impossible not to violate one or the other norm. But assuming

that ought-implies-can is correct, this could not be the case. One or the other presumptively

authoritative moral norms must not in fact be.

24 Finnis, “On Reason and Authority in Law’s Empire,” 380. See related in Dworkin, Law’s Empire, 110–114. 25 Finnis, Natural Law and Natural Rights, 28. 26 Kelsen, Principles of International Law, 149.

29

This result is significant, because it portrays one important aspect of the natural law

theorist’s conception of legal authority. The legal authority of a norm is content-dependent:

whether or not a law is legally authoritative depends on whether its content follows from (or is

at least consistent with) true moral claims. Thus, sociologically speaking, an immoral law

might have such-and-such origins, such-and-such linguistic structure, and agents might act as

though it has authority, while people go to jail for violating it, and on. But normatively (or

doctrinally) speaking, because the content of immoral law is inconsistent with moral truth—

because legal authority is moral authority, which is itself internally consistent—immoral laws

could not be authoritative.

Postivism and Legal Authority

At the other end of the theoretical spectrum, some hold that legal authority is of the same

kind as the authority of a contract or promise – not necessarily moral, but dependent on some

form of consent or social recognition. The theorist to whom most commentators ascribe this

view is H.L.A. Hart, though Kelsen and Anscombe will also be helpful in this connection.

Hart's view of authority is, in certain respects, not so dissimilar from the natural law view. Like

Dworkin and Finnis, Hart explicitly tells us that the difference between the gunman who

issues commands and the representative of a (legitimate) government who issues commands

is that the latter has the authority and the former does not.27

At the same time, the differences between Hart's view, what I will call the “positivist”

view, and the natural law view are as important as the similarities. On the relationship

between law and morality, Hart writes,

Those who accept the authority of a legal system look upon it from the internal point of view, and express their sense of its requirements in internal

27 Hart, The Concept of Law, 198–199.

30

statements couched in the normative language which is common to both law and morals: 'I (You) ought', 'I (he) must', 'I (they) have an obligation'. Yet they are not thereby committed to a moral judgment that it is morally right to do what the law requires.28

I take Hart's remarks here not simply to be about the kinds of beliefs that persons tend to be

committed to upon certain utterances, but what the content of those utterances logically

implies. Specifically, assertions about the authority of law, perhaps by a legal system itself, as

in the above quotation, are not assertions of moral judgments. Questions of moral authority

and questions of legal authority are, to use Hart's term, separate.29 Anscombe summarizes this

idea perfectly when she writes

Legal authority, legal validity, legal obligation: these are one thing; their presence or absence is to be ascertained by looking at certain institutions and their rules. It is another question altogether whether one should grant moral authority, moral validity, and moral obligation.30

Familiar cases of immoral laws and the possibility of civil disobedience support

Anscombe's convictions. The Jim Crow laws spring to mind immediately. Despite the fact that

these laws perpetuated and institutionalized morally abhorrent systems of belief, they were

the law. People followed them from a sense of legal obligation. It may be the case that, all

things considered (including moral facts) agents should not have followed these laws. But it

would be a mistake, I think, to dismiss, out of hand, the possibility that even immoral laws

such as these were genuinely authoritative. This, of course, is an example of domestic law.

Later in this chapter, I will discuss at length cases of specifically international law that make a

similar point.

One element of particular significance to the current conversation that comes out in

these considerations is that, on the positivist view, law might be defeasible. That is, it might be 28 Ibid., 199. 29 Hart, “Positivism and the Separation of Law and Morals.” 30 Anscombe, “On the Source of the Authority of the State,” 142.

31

the case that sometimes, when all reasons for action are considered, an agent ought not to

follow the law. As mentioned, certain outstanding instances of civil disobedience can be

perfect examples of when a person really ought to break the law. This does not make their

action legal. In fact, it is central to the very idea of civil disobedience that the action be illegal –

for it is that very illegality of an otherwise morally permissible (or obligatory) act that makes

civil disobedience a powerful political tool.31

But how do Hart and the others propose to make the leap from the fact that agents act

as though law has authority (including their behaviors, what they say, whether they comply,

how they respond to official censure, etc.) to the claim that law is genuinely authoritative and

not falsely so? Kelsen writes

Only human beings are capable of creating norms, that is, rules obligating and authorizing men. Insofar as human actions are supposed to be brought about by a psychic phenomenon called 'will,' norms are considered to be created by acts of will. The human will creating norms may manifest itself in acts consciously directed at the creation of norms, namely, in commands, in legislative acts, and the like; or in custom, that is, the habitual or usual course of acting accompanied by the conviction that men ought to act in this way.32

This is a familiar view, and one indebted to early modern social contract theory. Laws, one

might say, generally are authorized for their subjects because those subjects have, in one form

or another, willed to be bound by them. They accept the constraints that law or proposed law

would place on them, and in that act of acceptance authorize the law. This is perhaps mostly

clearly visible in the case of contract law. When I sign my contract with the university, both

parties take on new responsibilities. I will teach and do research for the university, and in 31 It is worth here mentioning that defeasibility is difficult to hold on a natural law view. If legal authority is moral authority, then the claim that law is defeasible is also the claim that morality is defeasible. But it is difficult to say that while morally, one ought to behave in one way, one ought to behave otherwise all things considered. What else is there to consider? While it is not so hard to think of cases where law is defeasible, it is harder to think of cases where morality is. 32 Kelsen, Principles of International Law, 150.

32

exchange, the university will provide a salary and health care. These actions are enforceable

as law, at least once both parties have given the appropriate signs of consent (a signature, a

handshake, etc.).

In contract law, parties to the contracts are the ones bound by its terms. In

international law, states are the primary objects of regulation. So just as individual consent is

necessary for the terms of a contract to be legally authoritative, Kelsen and others argue that

state consent is necessary for international law to be authoritative. The emergence of the treaty

as the primary and favored vehicle for international legislation speaks to this point. Unlike

custom (despite the UN's explicit inclusion of custom as a source of international law, and

Kelsen's insistence that custom is also an expression of consent), treaties almost always come

with clear indications of state consent (often both signature and domestic ratification).

Another significant aspect that this discussion of contracts brings out, an aspect that

again diverges from natural law, is that this positivist conception maintains that legal

authority is content independent. Law is authoritative because parties agree to it, but those

parties might agree to anything – even something immoral. But with regards to authority,

what matters is not the content, but the consent. Contrast this with natural law theory, which

demands that the content of law be consistent with moral truths in order to be authoritative.

Chapter 2 of this dissertation discusses the state consent view of international legal

authority in detail, so I will end this initial presentation with some summary remarks.

According to the positivist view, legal authority has at least the following three characteristics:

(1) defeasibility – authoritative law can be overridden by countervailing considerations (at least

moral ones) such that an action may be illegal while also the right thing to do, all things

considered; (2) willed – law depends on an act of will (however tacit) by its subjects in order to

33

be authoritative; (3) content-independent – law can be authoritative even if its content is

inconsistent with the best moral theory available.

Raz’s Conception of Authority

While the first two views dominate modern conversations about the authority of law, Joseph

Raz has also made significant and sustained contributions to the question of legal authority,

and has advanced his own account which is worthy of independent consideration.

Raz writes that “We should regard authority as basically a species of power, where

(normative) power is the ability to change protected reasons.”33 He defines 'protected reason'

as “both a reason for an action and an (exclusionary) reason for disregarding reasons against

it.”34 Raz provides the following helpful example: A mother tells her son to wear a coat, but

the coat is ugly, and the father tells the son to disregard his mother's command. The ugliness

is a reason not to wear the coat. The mother's command is a reason to wear the coat. The

father's instruction is an “exclusionary” reason, a second-order reason, that is a reason not to

consider the mother's instructions during practical deliberation. A “protected reason” is a

reason for action that is also an exclusionary reason to disregard countervailing reasons.

Power is the ability to change or, I assume, provide protected reasons, and authority is a kind

of power.

Raz’s attention to protected reasons is primarily motivated by two thoughts: First, that

authority is appropriately distinguished from advice and requests and second, that authority

is sometimes rightly defeated by countervailing concerns. Law is importantly different than

advice and requests since, in some sense, violating the law constitutes a wrong in a way that

ignoring advice does not. Raz differentiates advice from commands (like laws) by saying that 33 Raz, The Authority of Law, 18–19. 34 Ibid., 18.

34

commands are always “given with the intention that their addressees shall take them as

protected reasons,” whereas advice “is not [necessarily] given with an intention to be taken as

a protected reason.”35 So when I command you to wear that ugly sweater, I intend for my

demand to exclude other reasons you might have not to wear it. If I merely advise you not to

wear it, I do not necessarily have that intention. He writes:

If you request you submit yourself to the addressee's judgment on the balance of reasons, while at the same time trying to add a reason on one side of that balance. But one who commands is not merely trying to change the balance by adding a reason for the action. He is also trying to create a situation in which the addressee will do wrong to act on the balance of reasons.36

In advising, I provide new reasons to be compared with others the advisee already has. In

commanding, I supply new reasons in place of reasons held otherwise.

This difference between advice and commands has implications for how Raz

understands defeasibility. Like Kelsen, Hart, and other positivists, Raz agrees that law is

defeasible. That is, it may be the case that an agent ought not to follow the law, all things

considered. That said, the fact that law claims authority means that its commands

presumptively give exclusionary reasons – that is, they presumptively exclude certain kinds of

facts as reasons.37 A parent's command is defeasible by strong moral facts, but not by personal

fashion sense. State law is defeasible by federal law, but not by local custom. So for Raz, law is

still defeasible, but only by non-excluded kinds of reasons that he must specify.

What kinds of reasons does an authoritative command exclude? Raz is somewhat

cryptic on this point, but helpfully tells us that “what is excluded by a rule of law is not all

other reasons, but merely all those other reasons which are not legally recognized.”38 So moral

35 Ibid., 22. 36 Ibid., 24. 37 Ibid., 22. 38Raz, The Authority of Law, p.

35

reasons that have not been socially recognized as legal (perhaps by being used in previous

court rulings, or invoked in the preamble to important legal documents) are excluded from

practical deliberation about what to do. But moral customary and stipulative reasons that

have been legally recognized can still play this exclusionary role.39

This view, that authority is the power to change protected reasons – that is, that

authoritative commands are defeasible but only by certain kinds of reasons – allows Raz to

capture two intuitive aspects of legal authority: (1) it can be defeated, such that prima facie

illegal actions may not actually be, and (2) it takes the form of an authoritative command, in

contrast with mere advice or requests.

The question remains: What gives law this authority? The answer, what Raz calls his

“service conception of law,” says that law is authoritative when two conditions are satisfied:

First, that the subject would better conform to reasons that apply to him anyway (that is, to reasons other than the directives of the authority) if he intends to be guided by the authority’s directives than if he does not (I will refer to it as the normal justification thesis or condition). Second, that the matters regarding which the first condition is met are such that with respect to them it is better to conform to reason than to decide for oneself, unaided by authority (I will refer to it as the independence condition).40

Essentially, the view seems to be that the law has authority over an agent when that agent

would better comply with his or her own reasons for action by following the commands of law

than by following her own deliberative procedures. It is difficult, I think, to understand

“better” here in anything but moral terms, since Raz clearly agrees that convenience and 39 One problem to mark here, to which I will return later in this chapter, is that Raz does not offer much discussion about whether countervailing legally recognized rules suffice to make a prima facie illegal action legal or whether they make it the right thing to do, all things considered. It seems that by limiting the scope of exclusion only to legally recognized rules, the fact that a legal command is overridden by countervailing legal commands would make the action legal. But there are cases, like human right which I will discuss later, which are overridden by socially, but not necessarily legally recognized reasons. This broadens the scope of defeasibility beyond what Raz's view recommends, though the cases seem persuasive to me. 40 Raz, “Problem of Authority,” 1014.

36

desire do not suffice as justifications for legal authority.41 But Raz also says that legal

commands can exclude moral reasons, so it is unclear how legal authority could itself be a

source of moral reasons. This is only further confused, when Raz tautologically asserts that

“what is excluded by law is not all other reasons, but merely all those other reasons which are

themselves not legally recognized.”42

The details of Raz's view aside, the valuable insight we can take from him is the

importance of distinguishing between advice and legal command. Raz uses the idea of

exclusionary reasons to make this distinction, but struggles, I think, to successfully

incorporate this idea of exclusionary reasons into a larger conversation about the relationship

between law and morality. Still, the distinction is an intuitive one, and it should be accounted

for in any successful account of legal authority.

Recognition of International Legal Authority in State Action

Even given these existing views about the authority of law, I believe that there is reason to

persist in the investigation of what authority generally is. This belief stems not from

considered objection to these views (alone), but from the simple observation that none of

these, perhaps with the partial exception of Kelsen, develops a conception of international

legal authority for the international, rather than domestic, arena. Finnis, Hart, and Raz are all

interested in law within the state, particularly thinking about the relationship between state

power and individual autonomy. But the challenging questions to ask in the international

realm are very different from this. States are no longer simply sovereigns, but are instead

intermediaries between the global order and the lives of individuals. The subjects of domestic

law are people, the subjects of international law, peoples. Given these significant differences, 41 Raz, The Authority of Law, 22. 42 Ibid., 33.

37

it seems presumptuous to hold that accounts of legal authority settled in a domestic context

should also apply in an international setting. Some such view might well be defended, but it

would have to be defended in light of an interrogation of international law as we find it. I will

now turn to such an interrogation.

Any view that posits the authority of law is open to an early objection – one from the

descriptive or interpretive side of the discussion. International law, the anarchist’s objection

goes, does not exert any normative force on states. 43 If international law did exert such

normative force, states would at least sometimes comply with international law even when

doing so was not in their own state interests. However, because international law is

notoriously “toothless”, states rarely if ever comply with international law just because they

have a legal obligation to do so.44 Simply, given the vast amount of non-compliance apparent

in modern practice, state action cannot be faithfully interpreted as action guided by the law or

done from a sense of legal obligation. States may invoke international law as a justification for

actions, but this is only a veil for their true motives –- self-interest and the accumulation of

power.

But this is too quick. Simply because many states are selective about compliance,

because non-compliance is common, and because compliance frequently (or usually)

43 One might imagine such an argument coming from realist camp in international relations theory. See, for example, Morgenthau Politics Among Nations, and Waltz Realism and International Politics. Compare Chayes and Chayes 8. 44 One might point to the recent Russian involvement in Crimea for an example. Given the apparent unwillingness on the part of the US and NATO to become involved in further military conflict after the Iraq and Afghanistan wars, Russian leadership perceived an opportunity to violate international law with impunity. The Maidan protests over former Ukrainian president Yanukovych’s dealings with the EU were an opportunity to take advantage of this weakness, despite the illegality of interfering with the internal governance of a sovereign state. States only selectively comply with international law, and only then because it is in their interest.

38

coincides with state interest, it does not follow that states do not treat international law as

authoritative. In a similar vein, Chayes and Chayes write:

If national security regimes have not collapsed in the face of significant perceived violation, it should be no surprise that economic and environmental treaties can tolerate a good deal of non-compliance… As we have noted, a considerable amount of deviance from strict treaty norms may be anticipated and allowed for from the beginning, whether in the form of transitional periods, special exemptions, or limited substantive obligations, or by the informal expectations of the parties.45

Their idea, it seems, is that non-compliance is insufficient evidence both for the claim that

international law is not authoritative, and for the claim that international practice implicitly

denies the authority of international law.

So given the prevalence of non-compliance, what evidence is there in states’ actions

that they recognize international law as authoritative? To answer this question, it will be

helpful to distinguish some of the forms of compliance and non-compliance. As Chayes and

Chayes write, “Compliance is not an on-off phenomenon”46, by which I take them to mean

that there are different levels and forms of compliance. By understanding this more nuanced

conception of compliance, we will be in a better position to see how state actions are in fact

compliant, and how even truly non-compliant actions still implicitly recognize the authority

of law. For current purposes, I will distinguish six forms of compliance and/or non-

compliance: good faith compliance, strategic compliance, rogue action, simple failure, failed

proposal, and successful proposal.

The first pair, good faith and strategic compliance, shows that even perfectly compliant

actions can have significantly different normative characters. Good faith compliance occurs

when a state (or a state representative) acts in full accordance with international law, and that

action is motivated by a sense of legal obligation. Kant’s idea of actions from duty, as opposed 45 Chayes and Chayes, The New Sovereignty, 19. 46 Ibid., 17.

39

to actions in accordance with duty is a helpful analogy. Strategic compliance, by contrast, is

legal action motivated by state interest or some basis other than a sense of legal duty. Because

corporate agents are at issue, it can be difficult to tell what a state’s “true” motive is. With

many different representatives and internal disagreement, state action is hardly univocal.

Part of the realist’s objection seems to be that when we see compliance with

international law, we find only strategic compliance, and never good faith compliance. From

this, she wishes to conclude that even compliant actions are not performative assertions of the

authority of law. But strategic compliance does implicitly recognize the authority of law, since

part of a state’s strategic calculation to comply will include the political immunity and

normative high-ground granted by legal action. Compliant states can reasonably expect legal

actions to go unchallenged by the international community, which is itself an implicit

assertion of the normative priority of law in the international community.

The next pair, in this case two forms of non-compliance, also helps to illustrate how

even non-compliant actions can assert international legal authority. Rogue action is the

violation of the letter of international law with blatant disregard for the illegality of the action.

Such actions are perhaps the closest examples of state actions that do not implicitly recognize

the authority of international law.47 Again the use of chemical weapons in Syria can serve as a

helpful example. Although Syria had not yet ratified the CCW, one might reasonably believe

– given the international uptake of the prohibition against chemical weapons – that Assad’s

use of chemical weapons violated customary international law, and that this was sufficiently

evident to deem it a rogue action. 47 It is significant, I think, that despite the fact that rogue actions are probably the closest we can get to a state action that does not invoke the authority of international law, rogue actions still do so. One might argue, following Chayes and Chayes and Anne-Marie Slaughter, that state sovereignty is defined by international law (or at least a network of international norms for Slaughter), so any state action that uses the powers granted the state by international law (even rogue actions) is an implicit invocation of the authority of international law.

40

Rogue actions should be distinguished from simple failure to comply with international

law. Simple failure occurs when a state violates international law although there is substantial

evidence that the state did what it could not to do so. Given the aspirational character of much

of current international law, like the ILO’s labor standards and parts of the Universal

Declaration of Human Rights (UDHR), history is rife with suitable examples of states falling

short of the mark despite their good faith efforts. Unlike rogue actions, simple failures can

explicitly acknowledge the reason-giving capacity of international law, just as Aristotle’s

akratic man acknowledges virtue without being able to attain it.48

In both cases, states knowingly violate international law and nevertheless implicitly

assert its authority. Consider an analogy with driving. Drivers knowingly and willingly violate

speed limits. But usually those violations are minor, departing from the marked speed limits

by tolerably small amounts. If we think that agents only implicitly acknowledge the authority

of law when they comply perfectly, then speeders and drivers in slow and broken cars would

not be examples of such acknowledgment. But if we recognize that even these departures are

minor, and that, in general, they stay within tolerable limits generally expressed by speed

limits, then we can see how traffic law is even implicitly acknowledged in these cases. Traffic

law shapes deviant driving.

We can say the same about international law: It shapes deviance, such that departure

from the norm is an implicit act of recognition. For instance, by limiting the scope of

government violence to his borders, Assad implicitly recognized the boundaries of his right to

rule. Moreover, when he was accused of acting illegally, his immediate response was denial,

but this, in itself is an assertion of the authority of law. If Assad did not perceive international

law as binding, in some sense, then he would have no reason to deny that he had violated it.

48 Aristotle, Nicomachean Ethics, Book 7.

41

“Yes,” he might say, “International law applied to the state I represent and prohibits the action

that my state did. I violated international law. But what of it? It has no authority!” But this is

precisely what we do not see. Violators deny that they have broken the law, or deny that the

scope of the law includes their state or that particular action (as we see with Syria’s best

defense – that it had not yet accepted the CCW). What states do not do, or only very rarely do

in extreme cases, is deny that international law is authoritative.

The final pair of related state actions are failed proposal and successful proposal. In the

absence of a world congress, and during the development and emergence of a global judicial

system, much of the legislative and interpretive work of international law falls to states

themselves. This is transparently the case with treaties, where states are usually the parties

involved in drafting and negotiation. What is somewhat more complicated is the recognition

that even when states are not involved in explicitly legislative activities (like treaty

negotiations), their actions are, as one might put it, “proto-legislative”.

To say that state actions are proto-legislative, is to assert that state action should be

understood as a performative and interpretive proposal about what the law is, or, more

precisely, what the rights of states are as determined by international law. Because there is no

global sovereign, and because political power is decentralized in our world, state actions must

be judged as legal or illegal by other states; there is no other sufficiently powerful entity to fill

this role. When states act, they employ their right to rule. But this is not simply the exercise of

perfectly understood rights. State action is a performative interpretation of what the acting

state takes its rights to be. This proposal may be rejected by other states, resulting perhaps in

sanctions, public shaming, or even violent conflict if the action is radical and offensive

enough. Alternatively, the proposal might be accepted, allowed, and used as precedent for the

42

future actions of other states. Chayes and Chayes mark just this phenomenon when they write

that

It is, of course, by no means unheard of that states, like other legal actors, take advantage of the indeterminacy of treaty language to justify indulging their preferred course of action. Indeed, a state may consciously seek to discover the limits of its obligation by testing its treaty partners’ responses.49

While this passage lacks the ambivalence that Chayes and Chayes exhibit elsewhere about

such informal methods of legislation and interpretation, it suggests that state action can be a

kind of legislative proposal, a performative motion, as if the state were saying, “By my action I

propose that actions like this be considered legal.” In doing so, its hope is that other states will

follow their example and, in so doing, recognize that new right as part of what it means to be a

state.

For a clear example, consider the 2014 treaty between Russia and Crimea. By

recognizing Crimea itself, the Russian government performatively proposed to the world that

Crimea be recognized as an independent political entity (this appears to have been largely

successful). It was as if Russia were proposing to the world: “Here is something new that states

can do. States can make treaties with the Crimean government as if they were independent

from Ukraine”. Whether the statehood of Crimea becomes international law, sociologically,

will depend on whether other countries besides Russia decide to follow suit.

On a purely descriptive level, such proposals sometimes succeed and sometimes fail.

Failed proposals are ultimately violations of international law. Consider the analogous case of

civil disobedience. If one violates the letter of the law, doing so with the intention to bring

attention to the perceived injustice of that law, but fails ultimately to get the law changed,

then that person has simply violated the law. Successful proposals are less clear, and might

49 Chayes and Chayes, The New Sovereignty, 12.

43

actually be indistinguishable from compliance in certain cases. If what initially seems like civil

disobedience were accepted subsequently as the customary interpretation of the contested

law, then successful proposal is really just following the law – just according to an

interpretation that not everyone shared.50

While these six forms of state action with respect to compliance are almost certainly

not exhaustive, they create a picture of several kinds of ways that state action invokes the

authority of law, however indirectly. In some cases, states follow international law in good

faith because it is the law. But the authority of law is not acknowledged only by good-faith

compliance, since even non-compliance can tacitly imply the authority of international law.

After all, states are corporate agents, so their actions often have diverse purposes and the way

that they acknowledge international legal authority is more indirect. Sometimes states try to

propose new law, hoping that others will accept the change and augment the norms of state

behavior in their favor. Even when states break the law, often that failure is accompanied by

apologies, excuses, and justifications for an exception that equally well reflect that states take

themselves to be bound by international law.

The Conception of Legal Authority Implicit in State Action

State actions not only imply that international law is presumptively authoritative in modern

global politics, widely perceived trends in state behavior also offer some guidance about the

form of authority that states seem to take international law to have. More precisely, state 50 This analogy between performative proposals and civil disobedience is limited in that while state actions may often play this interpretive role, it is less obvious that the actions of citizens do the same. Part of why state actions play this role is that the international legal regime is still emerging, so that regime lacks many of the stable legislative organs that many domestic legal regimes have. Consequently, the potential for performative proposal in domestic law might be limited to minor reinterpretations of customary law, which are mutable by informal means, whereas the potential for performative proposal in international law ranges across all sources of law.

44

behavior with respect to international law indicates not only that international law is

authoritative, but that that authority has a specifiable form. In this section, I will argue that

the best interpretation of state action is one in which states act as though international law is

legally authoritative, which is to say that international law (1) gives states directed reasons for

action (2) is defeasible in light of countervailing considerations, and (3) is reason-giving

independently of its content.

Reason-Giving: Part of what it means for international law to be legally authoritative is that it

gives agents reasons for action. This characteristic of authority can be described in different,

but compatible ways. Raz, as we saw, offers some general insights into the reason-giving

nature of law.51 First, the fact that something is legal or illegal can suffice to conclude rational

deliberation. Second, the reasons given by law are different in kind from the reasons given by

requests and advice, a difference that can be explained in terms of protected reasons. I believe

that Raz’s characterization of law is substantiated by practice, though I think that there is a

simpler explanation of the difference between law, advice, and requests that is more faithful

to public understandings of international law. It seems uncontroversial to say that if I am

deliberating about what to do, and I am informed that one of the two options I am considering

is illegal, then that fact will bear on the ultimate decision I make. A second way to express

what it is for law to be reason-giving is to say that the legality or illegality of an option suffices

to end practical deliberation when all else is equal. So, again, if I am deliberating between two

options that stand equally before me in terms of moral cost, preference satisfaction, ease,

efficiency, and so on, and I discover that one of the two is illegal (though I know that this will

have no impact on the relevant costs and benefits that I can reasonably expect on either side),

51 Raz, The Authority of Law, 19.

45

then the mere fact that the option is illegal suffices to end my reasonable deliberation in favor

of doing the legal action.52

The US’s response to the beginning of what has now become the Syrian civil war is, I

think, helpful in this connection.53 The US did not send military aid to Syria. Given the US’s

fiscal and political situation in early 2011, this might have seemed like the obvious outcome.

Two long wars had stressed its economy and quashed any popular support for a new conflict.

But at the time, the possibility of military intervention under the aegis of the North Atlantic

Treaty Organization (NATO) seemed a real possibility. Opposition members in Syria were

almost univocally requesting external assistance, NATO had already played a role in a similar

conflict in Libya, and the presence and use of chemical weapons was a genuine shock to the

global conscience.

It would be a mistake to say that any one reason or any one event turned the tide

against US military involvement in Syria. It was surely a complex decision based on a delicate

balance of equally important and competing values. But one major moment during these

deliberations was the United Nations Security Council’s (UNSC) refusal to approve

intervention. With Russia’s veto, the UNSC’s decision rendered military involvement in Syria

illegal under international law (at least in the absence of special circumstances like the

emergence of a credible threat to the territorial integrity of another country). What is striking

though, and what speaks in favor of the presumed authority of law, is that this decision by the

UNSC made any difference at all. The political and economic situation before and after the

vote at the UN was unchanged. What did change was the legal status of intervention, and with

52 To say that law is authoritative only if it suffices to close deliberation when all else is equal is not to say that law cannot also close deliberation when all else is not equal. Thus it could be the case that even though acting legally will in fact result in significant costs for an agent, the law in question might be important enough to outweigh those costs. 53 Gritten, Rodgers, and Macguire, “Syria: The Story of the Conflict.”

46

it, the likelihood that Syria would see western military involvement. This, to me, is a clear case

in which the fact that something became illegal served as a reason, in itself, for an agent not to

do that thing.

This is an example of international law sufficing to end practical deliberation of states,

but does it show a distinction between law on one hand and advice and requests on the other?

It would seem to. When one goes against advice or chooses not to grant a request, one is not

normally presumed to become the appropriate target of accountability seeking behaviors.

Even when it is good advice, we generally presume that it is not wrong to choose not to take it.

But had the US sent military aid to Syria after the UNSC’s decision, this would almost

certainly have been the case. Apologies, justifications, and excuses would have been necessary

on the part of the US, which show that the UNSC’s decision changed the presumed normative

playing field from one where it was uncertain whether US intervention would have

constituted a wrong to one where it almost certainly would. This should not be a surprising

result. As I said before, Raz’s distinction between advice and law is an intuitive one, and a

successful account of international law will be able to explain it.

A second characteristic of the reason-giving capacity of law stands out in this example.

States that violate international law, or that choose to comply to avoid being held accountable,

are accountable to other agents. The US, had it contravened the decision of the UNSC would

have been accountable to the members of the UNSC and the states who it represents. This

presumption that compliance is, in a sense, owed to other agents is not unique to this case.

Signatories to a treaty are accountable for noncompliance to the other signatories. States subject

to customary law are presumed to owe conformity to other adherents of that custom. Margaret

Gilbert describes this phenomenon as the “directedness” of duties. Focusing on promises, she

writes

47

There is a well known connection… between directed obligation and owing…. In terms of the previous example: If Olive has a directed obligation to Roger to go to Chicago tomorrow, then, equivalently: Olive owes Roger her going to Chicago tomorrow… This construal of owing suggests a particular amplification of an important further point about promising… that by virtue of one’s status as a promisee one is in a position to demand the promised act from the pertinent promisor… Intuitively, a promisee is in a position to rebuke the promisor for non performance.54

Here, Gilbert’s idea is that when two parties enter into a promise, they become bound

together by a web of correlative rights and obligations. Promisors have duties to promisees;

promisees have correlating rights against promisors; promisees have the normative standing

to demand performance from promisors; and promisees are specially licensed to rebuke

promisors for non-performance. In many ways, the example of the UNSC and Syria is

isomorphic with Gilbert’s view of promises such that, like everyday contractual obligations

the presumed duty to conform to international law appears to be directed.

One final characteristic also stands out about legal reasons. While international law

occasionally addresses other kinds of agents, it is primarily addressed to states. The UNSC’s

decision determined whether intervention in Syria by states was legal or illegal. This makes a

certain kind of sense, since only states have the material resources necessary for such an

endeavor. That the reasons given by law are reasons for states is not limited to this case, but

apparently pervasive. Treaties are signed by and are presumed to bind states. The UDHR,

which we turn to next, is explicitly addressed to states as a specification of the duties they owe

their citizens. The ILO’s labor standards, which are importantly not a treaty but rather are a

form of administrative law, are explicitly addressed to states.55

54 Gilbert, “Three Dogmas About Promising,” 306–7. 55 International Labor Organization, “ILO Declaration on Fundamental Principles and Rights at Work.” “Declares that all Members, even if they have not ratified the Conventions in question, have an obligation arising from the very fact of membership in the Organization, to

48

Interestingly, recent years have seen the emergence of efforts to create international

law that addresses non-states. Legal frameworks to address terrorist organizations, for

example, might be an example.56 Likewise, the Accord on Fire and Building Safety in

Bangladesh — an document that bears a striking resemblance to better established examples

of administrative international law but for which it is hoped that corporations with factories

in Bangladesh, rather than states, will sign on — might be an example of international law

addressed to non-states.57 In the immediate post-war period, it might have been more

plausible to say that states were the exclusive addressees of international law, but that appears

to be changing. We can, I think, still confidently say that states are the primary addressees of

states. However, the superior account of international law will explain how international law

might also apply to secondary addressees like terrorist organizations, NGOs, and multinational

corporations.

Defeasible: While norms with legal authority suffice to end practical deliberation in some

cases, legal authority can also be overridden by countervailing considerations in others –at

least such seems to be the presumption implicit in current international legal practice.

Sometimes, given the specific contours of and balance of values at play in a specific situation,

the action demanded by law may come at a sufficiently grave moral or pragmatic costs that

disobedience is justified on the whole.

Human rights law is an excellent example of defeasibility. Human rights, particularly

those specified in the UDHR, are generally taken to be an instance of international law and

are followed from a sense of legal obligation (as well, perhaps, a sense of moral obligation). At respect, to promote and to realize, in good faith and in accordance with the Constitution, the principles concerning the fundamental rights which are the subject of those Conventions,” 56 Asad, On Suicide Bombing; Mamdani, Good Muslim, Bad Muslim. 57 International Labor Organization, “Bangladesh Accord.”

49

least, when they are followed, they are followed from this sense. But many of the articles of the

UDHR demand that governments provide services that go beyond their material capabilities.

Consider Article 26, which guarantees every person a free elementary education. For many

countries, the infrastructure, training, and general institutional upheaval necessary for

compliance is Herculean at best and practically impossible. So given the sundry exigencies of

political life in much of the world, and given the practical impossibility for governments in

many of these places to institute such a system of free elementary education, is it wrong,

overall, for a government to fail to comply? My sense of the international community is that

no, they do not. Are these countries in violation of the law? Yes. Is this wrong of them, on the

balance of moral, legal, and prudential interests? The international community seems not to

think so. We do not, for example, see representatives at the UN mulling over sanctions against

the Philippines because of the suspension of public education during Haiyan. Nor do we see

countries publicly criticizing the Burundian government on its problematic education system,

because we understand that they have bigger problems to attend to first (like its alarming

levels of food insecurity).58 In such cases, international law is still authoritative. It is simply

overridden by countervailing concerns.

Considering these cases, though, one might object that in fact, the fact that countries

can fail to comply with international law and nevertheless avoid sanction means that the law

is not authoritative at all. I think there is reason to disagree. First of all, I have already

discussed, earlier in this chapter, how even non-compliance can still tacitly acknowledge the

authority of law. Second, and specifically with respect to this example, we do see international

pressure for compliance in the cases of countries that do not have countervailing concerns the

way that the Philippines and Burundi do. Saudi Arabia, for example, almost certainly has a far

58 World Food Programme, “Hunger Statistics.”

50

better developed primary education system them Burundi. Nevertheless, it has been subject to

various forms of “soft coercion” by UNICEF and the UNDP and other international regulatory

bodies.59 Whatever the reasons are – whether recognition of economic differences between

the countries, a substantive moral objection to Shar’ia, or relevant political exigencies – the

case seems to be that the laws that guarantee elementary education to all are indeed

authoritative, they are just overridden in specific cases.

Content-Independence: To say that legal authority is content-independent is to say that law can

be authoritative independently of the relationship between its content and generally accepted

or otherwise true moral claims. Put simply, international legal practice seems to reflect the

belief that non-moral and even immoral laws can bind states. The most common examples of

content-independence are morally arbitrary laws. The locations of borders, for example, are

almost always morally arbitrary as cosmopolitan theorists of global justice are wont to remind

us.60 Nevertheless, the laws ensuring those borders, however morally arbitrary they may be,

are legally authoritative.

Again, one might object to this point by arguing that while specific borders may be

morally arbitrary, the laws protecting borders in general are not morally arbitrary because, for

instance, borders help states avoid violent conflict by harmonizing and coordinating public

understandings of the scope of state power. Where borders are may be arbitrary, but that states

have border is not morally arbitrary. Therefore, even apparently non-moral or morally

arbitrary laws are authorized by true moral claims.

But this objection is more difficult to sustain for cases of immoral laws that are

nevertheless treated as authoritative. Consider the North America Free Trade Agreement 59 Hamdan, “Women and Education in Saudi Arabia.” 60 Rawls, “The Law of Peoples”; Pogge, “An Egalitarian Law of Peoples,” 465.

51

(NAFTA). In 1994, the US, Canada, and Mexico agreed to gradually lower economic trade

barriers between their countries.61 This process was completed in 2008. The hope was that

trade liberalization would improve per capita income and quality of life for members of all

countries, but in fact we are seeing unexpected and morally problematic consequences

emerging from the deal. Perhaps most striking among these is that Mexican corn and dairy

farmers are being displaced from their land because of a sudden inability to compete with

(subsidized) foreign crops. Because NAFTA does not take care to correct for such externalities,

the law simply and straightforwardly harms these people. Despite the economic and political

advantages promised (and even delivered) by NAFTA, it is not obvious that these suffice to

save the law from these fairly damning moral objections. Nevertheless, the US, Mexico, and

Canada refrain from introducing new economic barriers to trade that might protect

agriculture in Mexico and other vulnerable industries. NAFTA contradicts morality, but it is

still acknowledged by all parties as legally authoritative.

The Idea of Legal Authority

From these suggestive examples, we can distill the idea of legal authority presumed in

international practice. International law, and perhaps law more generally, gives states reason

for action. Cases such as the US’s choice not to become involved militarily in Syria after the

UNSC vote show that law can suffice to end well-reasoned practical deliberation – it can tip

the scales without any further evaluation or judgment by the agent. These given reasons are

presumed different from the reasons given by advice and requests because, unlike refused

requests or ignored advice, violations of law constitute a wrong of some kind. These reasons

for action, the duties or obligations that states have to comply with international law, are

61 Teslik, “NAFTA’s Economic Impact.”

52

directed. Compliance is presumed to be owed to other states. Although states are the primary

addressees of international law, and therefore the primary actors affected by these reasons,

international law might also apply to some secondary addressees including NGOs, terrorist

organizations, and multinational corporations.

While the legality or illegality of some proposed action can suffice to end practical

deliberation, it does not always do so in practice. Cases like the widespread violation of

human rights law show that international law and its authority can be overridden by strong

countervailing considerations. Even though law demands that governments provide their

people certain basic primary goods like education, some countries simply lack the material

resources to do so. What we find in practice is that countries that genuinely lack the material

resources to provide these goods, or otherwise can provide convincing reasons that excuse

their non-compliance, experience far less political and legal pressure than countries with

these resources, even when legally required social programs are comparably limited. In a

sense, the international community forgives or excuses such violations in recognition of the

practical limitations of governments while nevertheless maintaining that such human rights

deficiencies are illegal under international law. In a word, international law is presumed to be

defeasible by sufficiently grave countervailing concerns.

Finally, we find that the presumption of law’s reason-giving capacity and defeasibility

persists independently of the content of international law and its relations to true or widely

accepted moral claims. Even immoral laws are presumed to give states reasons for action, and

in this way, legal authority is presumed to be content-independent.

These central cases of the practice of international law support the identification of

three characteristics that are necessary elements of legal authority — directed reason-giving,

defeasibility, and content-independence. At least, an explanation of these characteristics will

53

be a central and important part of any theory of legal authority aimed at understanding and

justifying modern practice.

If what I have said is correct and international legal practice supports my

characterization of these three characteristics of legal authority as we find it in the global

context, then the question is: How could a legal order justifiably bear this kind of authority? I

turn to this question next.

54

CHAPTER 2: State Consent

Introduction

In Chapter 1, we discussed four accounts of authority, arriving finally at a conception of

authority derived from an interpretation of going international legal practice. To summarize,

when participants in international politics engage in activities that they themselves take to be

governed by legal rules, however obliquely, their actions implicitly presume its authority. An

interpretation of those actions can identify the kind or form of authority presumed, and I

argued that legal authority as presupposed in practice has at least the following three features,

what we may call its authority characteristics.

REASON GIVINGNESS: Law gives states reason for action such that (1) they owe their

performance to other states, and (2) these reasons are distinct from advice and

requests.

DEFEASIBILITY: The obligation to comply with international law can be overridden

by sufficiently grave countervailing concerns; and

CONTENT-INDEPENDENCE: Whether a law is authoritative or not is not a function

of the morality of its content.

But of course the fact that practitioners take law to have this kind of authority does not

make it so. The mere fact that an agent believes or accepts some propositions is insufficient

evidence to justify the further claim that the propositions in question are themselves true.

After all, epistemic agents are fallible, whether individual or corporate. So, assuming that the

goal of this project is something like to offer some argument for the actual authority of

international law, more must be said.

Even so, the foregoing analysis of presumed legal authority remains of great value to

us. It would be a mistake, for a project aimed at justifying international legal obligations, to

55

take these purely sociological facts as normative justification. They are not the justificans. (Or

at least I will not be among those who treat them as such, as Austin, for example, might,

however fruitful this may be for certain projects.) A different and philosophically more

interesting project is to take the form of authority presumed in practice to be the justificandum.

As we saw in Chapter 1, other theories so far have started from armchair conceptions of legal

authority (or, at best, conceptions designed to suit the domestic context), and then asked how

international law might come to meet the criteria necessary or sufficient to have that kind of

authority. One might plausibly speculate that this methodology is partially to blame for

enduring disagreements in legal theory. Rather than starting from an armchair conceptions of

authority – ones that, perhaps, are never even imagined by practitioners – I am proposing that

we begin from the conception of legal authority that participants in international political life

actually seem to use and invoke in their activities, and then see whether any argument can be

given for international law’s indeed having that form of authority.

To that end, the question of this chapter and the next is whether international law as

we find it is really is authoritative, in the interpretive sense specified, and if so, how this might

be so. This chapter will be dedicated to one answer, the state consent argument, and the next

to a different answer, Dworkin's New Philosophy.

The discussion in this chapter will proceed as follows. First, I will offer an argument for

two criteria of success in any attempt to justify the legal authority of international law as we

find it. Any successful answer must satisfy two criteria:

FORM CRITERION: what is justified must have the kind or form of authority

interpretively implicit in the practice.

SCOPE CRITERION: all or at least the preponderance of rules respected as

authoritative international law must be accounted for.

56

I will argue that, while an appeal to state consent presents the form of authority we are

looking for, it would force us to deny the presumed authority of a large subset of current legal

rules. Large swaths of law, perhaps even including treaties, are not consented to – explicitly,

tacitly, or hypothetically. The failure of state consent theory to meet the scope criterion then

raises the question whether some less-revisionary position is available, a question I take up in

later chapters.

Form, Scope, and Explanatory Success

If legal scholars agree about anything, it is that existing legal systems are open to constructive

criticism. Certainly existing legal systems are not perfect, and there is value in prescribing

corrective measures. Yet this assumption discourages us from asking questions about how

our current international legal system, or something like it, might actually be justifiable at all.

Unlike questions motivated by a desire for change, this question is motivated by an optimistic

conservatism — a sense that calls for revolution are premature, and that a specific, immanent

form of critique may be more powerful. We already know, presumably that our world is not a

utopia. But it would be quite another matter to find that our legal system, and anything like it,

could not be what it purports to be, and that its practioners suffer from some kind of grand

illusion.

In Chapter 1, I argued that the body of rules recognized by states as international law is

presumed to have a specifiable form of authority and presented the fundamental question of

this project:

THE BASIC QUESTION: How could international law have the kind of authority it is

presumed in contemporary practice?

Any theory meant to answer this question must satisfy two criteria. First, it must offer some

57

justification for international law that explains how law could have the three authority

characteristics specified earlier. Second, given the scope criterion, that justification must

apply to all or at least the preponderance of existing law.

But why, one might wonder, should international law have any presumed kind of

authority at all? People presume all sorts of crazy things, and so why take the presumption of

practice to be an indicator of truth or anything of interest to anyone outside of sociology?

Putting the objection more concretely: because there is nothing special about what practice

presumes per se, the discovery that international law might genuinely be authoritative in the

way it is presumed to be would not illuminate any interesting philosophical questions. Putting

the objection more bluntly: Why is the basic question worth asking?

There is independent reason to pursue a theory that not only non-revisionary about

the scope of law, but also about its form. Morally speaking, it would only be necessary to

abandon current practice to the extent that either the scope or the form of law cannot be

justified somehow. Such social upheaval may be difficult and costly, fraught with false starts,

political resistance, and the threat of undermining the parts of society that are not

objectionable – in the proverbial throwing of the baby out with the bathwater. Moreover,

while the current international legal order may be objectionable in certain local issues, it does

not seem rotten to the core. Our world has made significant steps in the past century towards

securing basic human rights, economic prosperity, and global security, many of which have to

do with the establishment and maintenance of the emerging body of international law. These

points at least suffice to justify asking the question whether international law as we find it in

both form and scope might be justified. It may be the case that no argument can be given for

this claim, but there seems, to me, to be both reason to look for such an argument and reason

to hope that that search will not be futile.

58

State Consent Theories

The state consent view is perhaps the most familiar account of the authority of international

law, if only because, at least lately, it is also the most roundly rejected. Presenting the view,

Kelsen writes

An essential element of this view is the idea that the international community, or the legal order constituting this community, i.e., general international law, is based on the common consent of the states, or, which amounts to the same, on a contract of the members of this community.62

The basic idea behind the state consent view is that international law has authority for states

because they have in fact previously consented to comply with it. International law, according

to the theory, is akin to a contract, which is binding only for participants, on whatever the

parties can agree to.

Treaties and multi-lateral conventions are almost certainly the paradigmatic examples

of international law on this model. In the standard case, treaties are generally responses to

crises or potential crises in international politics. This crisis might be a military conflict, the

emergence of an environmental threat such as global warming, general concern about the

existence and potential use of particularly terrible weapons, or any of a number of other

events. Whatever the specifics may be, these conditions align the interests that states have in

coordinating their behavior to resolve or prevent the crisis. States send representatives to

discover, negotiate, and enumerate mutually acceptable terms for this coordination. This

done, states communicate their will to comply with the negotiated terms, usually by signing

the treaty and then ratifying domestically.

Two arguments can be mobilized in favor of state consent theory. The first emphasizes

62 Kelsen, Principles of International Law, 153.

59

the sovereignty of states. States, the argument begins, are sovereign. Sovereignty, by definition

and tradition, includes the right to autonomous self-governance. Generally, this right is

presumed to include the rights to control borders, to raise and control a military, to extract

and sell resources from owned territory, to pass and enforce laws domestically, and perhaps

more. If a state is forced to act against its own will, in any of the above mentioned areas

covered by state sovereignty, perhaps by another state or by some other global power such as

a terrorist organization, then this violates its right, unjustifiably. However, if a state agrees to

act according to the terms of some negotiated deal, such as a treaty, then that state has an

obligation to other signatories to do as it said it would. This is true not only of treaties, but also

whenever it can be rightly said of a state that it consented to what would otherwise be an

unjustified infringement of its right to self-determination. Thus even customary and

administrative law might be justifiable, if we can rightly say that states consent to them. States

therefore have an obligation to comply with international law – that is, law is authoritative for

them – but only, it is maintained, when states consent to this.

The second argument gives more attention to why it is that states have the right to self-

governance in the first place, by focusing on the rights of individual citizens within states. One

might be skeptical that states, which are artifacts and tools created by human persons, are

endowed with any rights at all. Rights are held by persons, one may claim, but not by

corporate entities like states unless granted to them by persons. Thus, the familiar argument

goes, states do not have a right to self-governance unless they are granted that right through

some sufficiently democratic domestic process. Persons must express their wills to transfer

their natural right to self-governance to their state. And, likewise, the state must give consent

to transfer its second-hand right to the international legal regime. Assuming that any state

action must reflect the will of its citizens, however indirectly, the consequence of this line of

60

thought is that if a state is forced to do some action that it does not consent to, then (at least

some of) its citizens are being forced to do something that they have not consented to, in

violation of their personal right to self-determination.

This theory has clear roots in the social contract tradition most closely associated with

authors like Hobbes, Locke, and Rousseau, but it’s history reached much further. Brierly

writes

The doctrine that consent may be the basis of legal obligation is at least as old as the Digest, where Hermogenianus is quoted for the proposition that rules which have been approved by long custom and observed for very many years, are observed no less than those that are written, velum tacita omnium conventio; and it would be possible to collect an imposing array of authority to a similar effect from the subsequent literature.63

It also resonates with prominent 20th century authors in the legal positivist tradition. Hart’s

discussion of “mutual restriction” as the grounds of political obligations, for example,

exemplifies the same commitment to voluntarism.64 Kelsen’s skepticism towards natural

duties also resonates, like when he writes

Legal principles can never be presupposed by a legal order; they can only be created in conformity with this order. For they are “legal” only because and insofar as they are established on the basis of a positive legal order … Certainly the creation of substantive law is not a creation out of nothing. Legislation as well as custom is directed by some general principles. But these principles are moral or political principles, and not legal principles, and consequently cannot impose legal duties or confer legal rights upon men or states as long as these principles are not stipulated by legislation, custom, or treaties.65

We also see similarities between the state consent view and other contemporary conceptions

of political obligation. Margaret Gilbert’s joint commitment theory, for example, shares

Kelsen’s sense that political obligation is best understood in terms of the social relations 63 Brierly, The Basis of Obligation in International Law, 9.Brierly 9 64 Hart, “Are There Any Natural Rights?,” 185; Also see Arneson, “The Principle of Fairness

and Free-Rider Problems,” 616. 65 Kelsen, Principles of International Law, 150.

61

between agents rather than universalistic moral theories.66 She writes

I believe that there is a central kind of obligation other than moral obligation such that if one is so obligated, one has sufficient reason to fulfill the obligation. In focusing on a “demoralized” problem of political obligation, I do not mean to downplay the importance of the following question: Does one have a moral obligation to support the political institutions of one’s own country in particular — and if so, what precisely is the ground of such obligation? This, or something like it, is the way the problem of political obligation is generally couched in terms of moral obligations. I hope to show that much is lost if the only version of the problem considered is moralized in this way.67

Despite a venerable lineage and a few sympathetic proponents, state consent theory has

become something of the whipping boy of contemporary international legal theory, and has

been for the last century or so. In 1933, Lauterpacht argued that the view is self-contradictory.

The view depends on the assumption of the legal principle pacta sunt servanda, whether or not

states accept that principle, while also denying the existence of such unconsented legal

norms.68 Decades later, Kelsen, who had sympathetic philosophical leanings to the state

consent project, argued that state consent was at odds with the empirical facts about

international law, that states do and have not actually or tacitly consented to international

law, and that state consent poses peculiar problems for how to understand the obligations of

newborn states.69 By the publication of D'Amato's book The Concept of Custom in International

Law in 1971, the state consent view had already become the view that any “serious author” had

to reject, but that nobody would seriously defend except “Soviet jurists who have seized upon

the notion of strict consent as a way to either reject 'capitalist' norms or simply, in Lissitzyn's

words, 'to pick and choose among the norms of international law'.”70 Thus swung the

pendulum between positivists and natural law theorists up to the present time. While there 66 Gilbert, A Theory of Political Obligation. 67 Gilbert, “De-Moralizing Political Obligation,” 390. 68 Lauterpacht, The Function of Law in the International Community, 417–420. 69 Kelsen, Principles of International Law, 152–155. 70 D’Amato, Concept of Custom in International Law, 188.

62

are still some willing to defend the state consent view, they seem to constitute something of a

minority.71

But state consent theory should be taken seriously. After all, the widely cited ruling of

the World Court in the Lotus Case upheld the state consent thesis and installed it as legal

precedent when it held that

international law governs relations between States. The rules of law binding upon States therefore emanate from their own free will as expressed in conventions or by usages generally accepted as expressing principles of law and established in order to regulate the relations between these co-existing and independent communities or with a view to the achievement of common aims.72

Given the general methodology of deference to practice preferred in this dissertation, and

assuming that this important case is indeed representative of practice, it seems that state

consent is at least worthy of consideration. Nothing I will say will vindicate the state consent

view—Kelsen arguably did away with any chance of doing that sixty years ago.73 But I will

suggest that, like a canary in a coal mine, even its failure is instructive, particularly for our

questions about the nature and basis of authority in contemporary practice.

State Consent and the Form Criterion

I have proposed two criteria for any successful answer to the basic question how international

legal rules recognized in practice might genuinely have the kind of authority they are

presumed to have. Part of why one might think that the widespread rejection of the state

consent view was premature is that this theory perfectly satisfies the form criterion.

Reason-giving: As discussed earlier, the reason-giving authority characteristic has three

71 See, for example, Rocheleau, “State Consent vs. Human Rights as Foundations for

International Law.” 72 S. S. Lotus (Fr. v. Turk.) (PCIJ 1927). 73 Kelsen, p. 154-5

63

necessary and collectively sufficient elements. First, a legal rule is reason-giving only if it can

suffice to conclude practical deliberation for states in normal circumstances about whether to

comply. For the state consent view, the international legal order is essentially a global contract

between states or at least it is made up of many of smaller contracts including treaties and

conventions. Generally, theorists agree that agents have sufficient reason to keep their end of

an agreement, though they disagree about why. I discussed two such explanations at the

beginning of this chapter. But assuming that agents do give themselves sufficient reason to

perform by consenting to the terms of a contract, and assuming that states are just like any

other agent with respect to this fact, if states do in fact consent to international law then it

appears to give them such conclusive reason.

Second, a rule is reason-giving in the relevant sense only if it gives directed reason for

action. Again, the analogy with agreements is instructive. As Gilbert correctly says, promises

and agreements give directed duties and rights.74 These are directed reasons for action —

reasons that explicitly make reference to what one owes to others or is owed by them. If

international law is a kind of agreement or contract between states, and the duty to comply is

a contractual duty, then that duty is owed to the other contracting parties and correlates with

their right to conformity against others.

Third, a rule is reason-giving in the sense relevant for authority only if it can be

distinguished form the kinds of reasons given by advice and requests. According to the state

consent view, violating international law is wrong in the sense that it violates an important

practical norm — likely something to do with trustworthiness or keeping one’s word. Gilbert,

for instance, thinks that violating the terms of an accepted contract in otherwise normal

74 Gilbert, “Three Dogmas About Promising”; Gilbert, “Is an Agreement an Exchange of

Promises?”.

64

circumstances contradicts standards of rationality.75 Without necessarily saying that it is

morally wrong to violate an agreement — it is probably morally right to violate an immoral

promise — we can nevertheless agree that it is wrong in some sense. Since the state consent

view makes legal duties a kind of contractual obligation, it can use this analysis to explain why

it is wrong to violate international law in usual circumstances. By contrast, there is no wrong

at all in ignoring advice or rejecting requests in normal circumstances. Perhaps it is impolite

or imprudent to do so, but not wrong in such a way that a person would become the

appropriate target of rebuke. Thus the state consent-view can distinguish these different

forms of reason-giving.

Defeasibility: Practice presumes that legal rules suffice to conclude practical

deliberation in normal circumstances, but also that those rules can be defeated by

countervailing considerations in special situations. It is never legal to break the law, and that

counts for something, but sometimes one ought to break the law all things considered. Again,

the analogy with contracts helps make sense of this. One usually ought to keep agreements.

However, if an agent makes an agreement and subsequently discovers that performance

demands the violation of a significant moral rule, or is incompatible with a pre-established

commitment, or even demands extreme personal risk or harm, then one probably should not

perform. International law is just another agreement, perhaps prioritized above other

agreements in virtue of the vast impact it can have on persons’ lives but not because legal

obligations are different in kind. And like any other contract, it is at least conceivable that a

situation could arise such that, although the contract was still in full effect and had not been

excused or completed or otherwise terminated, all things considered, one ought not perform.

Content-Independence: Finally, practice presumes that whether a legal rule is binding for

75 Gilbert, “De-Moralizing Political Obligation.”

65

states is not a function of that rule’s content. Of the authority characteristics, this, I think, is

the most controversial for the state-consent view. Do immoral agreements bind? If the state

consent view is going to capture the form of authority presumed in practice, that theory must

either claim that immoral agreements do bind, or claim that there is something special about

agreements between states that socially excuses them. Because the state consent view relies so

heavily on the analogy between agreements between individuals and agreements between

states, the latter option seems perilous.

The contention that immoral promises do in fact bind is not uncommon, but

philosophers certainly have not reached consensus on this point. Here Hobbes is ambiguous

on this point. He claims that covenants that violate standards of self-care are void but also

allows covenants extracted coercively.76 Gilbert is less ambiguous on this point. She prefers

what she calls Pritchard’s Point — “Once call some act a promise, and all question of whether

there is an obligation to do it seems to have vanished”77 — to the dogmatic insistence that

immoral promises do not bind.78 As support for this preference, she writes, “…if what seems to

be an immoral promise has not been kept, it can intelligibly invoked by the promisee as a

basis for complaint against the promisor.” This, she takes to be evidence that immoral

promises can be made, but separates that from the question whether they obligate. On that

point, she writes:

I propose that what is intuitive is this: promising is a source of obligation such that if one has made a promise one is in some sense obligated to do what one has promised, all things considered. In particular, a promise obligates irrespective of its content.79

If what Gilbert has said about promises is correct (and I think it is), and if promises are in 76 Hobbes, Leviathan Book I Ch. 14. 77 Prichard, Moral Obligation and Duty and Interest, 198. 78 Gilbert, “Three Dogmas About Promising.” See also Anscombe, “Rules, Rrights, and

Promises.” 79 Gilbert, “Three Dogmas About Promising,” 304.

66

relevant respects like agreements,80 and if international law is a kind of agreement between

states, then international law binds states, as Gilbert says, “irrespective of its content”.

I have used the work of Margaret Gilbert to argue that the state consent theory, if

correct, can explain and justify the presumptive form of international legal authority. To my

knowledge, this represents a novel application of elements of her more general plural subject

theory of joint commitment, though I will not here go into the details of that view.81 I should

be clear: to my knowledge, the state consent account of legal authority is not a view to which

Gilbert has committed herself, nor it one to which published views commit her, though I

believe it is consistent with her wider contributions to social and political philosophy.

If what I have said is correct, the state consent theory could justify the form of

international legal authority presumed in practice. The state consent view reimagines the

international legal order as a (social) contract between states. By limiting their own sovereign

rights through expressions of willingness, states take on rights against other consenting

parties that they comply but also undertake duties to the others to do their part. By

supplementing this basic thesis with details from Gilbert’s theory of agreements and

promises, we can see how law might give states directed reasons for action, be defeasible, and

bind states independent of its content.

State Consent and the Scope Criterion

While state consent theory might be able to satisfy the form criterion, it cannot satisfy the

scope criterion. If for no other reason, this is because there is no available interpretation of

going international legal practice in which all states consent to all of the international legal

rules presumed to bind them. This is not a new objection. In fact, as far as I can tell, this is the 80 Gilbert, “Is an Agreement an Exchange of Promises?”. 81 Gilbert, On Social Facts; Gilbert, A Theory of Political Obligation.

67

most influential objection to state consent theory, though it has a variety of versions. Hart, the

leading philosopher of legal positivism, writes

…it is submitted that there is no basic rule providing the general criteria of validity for the rule of international law… It is true that, on many important matters, the relations between states are regulated by multilateral treaties, and it is sometimes argued that these may bind states that are not parties. If this were generally recognized, such treaties would in fact be legislative enactments and international law would have distinct criteria of validity for its rules. A basic rule of recognition could then be formulated which would represent an actual feature of the system and would be more than an empty restatement of the fact that a set of rules are in fact observed by states. Perhaps international law is at present in a stage of transition towards acceptance of this and other forms which would being it nearer in structure to a municipal system.82

For Hart, the problem is not that states do not consent to or otherwise accept individual rules

to which they are subsequently bound. Instead, the problem is that states do not generally

accept any secondary rule of recognition that would suffice to bind some states to certain

rules, even when they have not consented to them specifically. Thus while rules that govern

interactions within a society of states, they do not comprise a proper legal system. This has the

dual implications that (a) nothing that states presume to be international “law” really is

genuine law, and (b) states that do not consent to treaties and other supposedly legal

arrangements are not bound by them, even though they are frequently presumed to be.

Kelsen, one of Hart’s major interlocutors and the person to whom Hart takes himself to be

responding in the above passage, writes

But actually men are by their nature neither free not equal; and even if they were free and equal by their nature, they are, by law, bound to behave in a certain way, without their consent. The theory of a common consent of the states voluntary entering the international community, as the basis of this community or of the international law constituting this community, rests on the same fiction. The states are

82 Hart, The Concept of Law, 230–231; Dworkin’s arguments against the state consent view seem

also to follow this train of thought. See Dworkin, “A New Philosophy for International Law,” 6–10.

68

bound by general international law without and even against their will. Thus, for instance, a new state, as soon as it comes into existence, has all the rights and all the duties stipulated by general international law, without any act of recognition of general international law on the part of this state being necessary.83

Even Kelsen, who Hart casts as the last, best, positivist defender of the authority of

international law, rejects the state consent thesis. His specific objection is different from Hart’s

however. For Hart, the failure of international law is that states do not consent to a secondary

rule of recognition. For Kelsen, it is that specific states do not consent to specific (primary)

rules, yet are bound by them.

What is of interest to the current project is not that the state consent thesis fails, but

how it fails. According to Kelsen’s diagnosis, the problem with state consent theory is that, in

practice, each state does not consent to each rule. From this we can infer his vision of the state

consent thesis: every state must consent to every individual rule to which it is held. This is a

very strong thesis — so strong that I doubt whether many philosophers would defend it

seriously. But according to Hart’s vision of the state consent thesis (which, remember, he

rejects), each state would not have to consent to each rule. Rather, states would only have to

accept a secondary rule of recognition. That they in fact do not is the real problem for state

consent theory.

What both of these authors assume, I think rightly, is that any normative theory of law

has to be to some extent responsive to the facts on the ground. What both Kelsen and Hart are

trying to draw our attention to, and the basis of their respective critiques, is that the state

consent thesis, if correct, would demand vast and sweeping revisions of current practice.

Customary law would be out. Administrative law would be out. Any multilateral treaty or

83 Kelsen, Principles of International Law, 154; See also Dworkin, “A New Philosophy for

International Law,” 8, where Dworkin raises a similar objection, though he does not cite Kelsen.

69

convention (including the very treaty of Westphalia) would not be binding for states that did

not exist at its creation without some explicit sign of acceptance. And this is a problem for state

consent theory, Kelsen and Hart say.

Why is being highly revisionistic a problem? After all, everyone agrees that the current

international legal system has problems that demand revision. Sure, states act as if what they

call international law is authoritative, but humans and the collective agents they constitute are

fallible. What, in principle, is wrong in holding that they are simply wrong about much, most,

or all of law? The problem is that such vast revisionism represents a fatal break between

philosophy and the lived world. There is something we do, in our world, called international

law. It appears to be normative. It is widely accepted. It shapes behavior and high stakes

decisions for real people every day. That such a system exists in a sociological sense is

undeniable. What Hart and Kelsen can teach us is that we should be doing work about that

system; that if anything is genuinely normative, then so is this; and that if we have a legal

theory that teaches us that the entire system is, or even just could be, a mistake, such

incompatibility should count against the theory, not against the genuine normativity of the

practice. Dworkin, I think, misses precisely this point when he writes

Hart himself raised the question whether so-called international law really counts as law on his new test. However, though he phrased that question in the traditional way, he actually changed the subject. He asked a question for social scientists…84

It might be overly charitable (even self-indulgent) to claim that something like the argument I

have suggested here is what Hart or Kelsen had in mind. But neither author was unclear about

their interest in Dworkin’s doctrinal question, and it is certainly uncharitable to say that they

are simply talking past it.

84 Dworkin, “A New Philosophy for International Law,” 4.

70

In Defense of State Consent

There is one, and perhaps only one, way to save state consent theory from this objection: we

can lower the bar for what counts as consent. The view that Hart and Kelsen (particularly

Kelsen) critique seems to assume that the kind of consent necessary to authorize legal rules is

explicit. Handshakes, signatures, or the utterance akin to Hobbes’s oath: “I authorize and give

up my right of governing myself to this man, or to this assembly of men, on this condition, that thou

give up thy right to him, and authorize all his actions in like manner.”85 But perhaps these explicit

signs of consent are not necessary. After all, informal agreements between friends (informal,

but still agreements with all attendant obligations and rights) can be accepted with a nod, a

smile, and perhaps even silence in the proper context. Even if each state has not given explicit

consent to each international legal rule by which it is bound, perhaps they have given consent

in one of these watered down senses. If we could offer an intuitive argument for other forms

of consent, and if we could show that states have engaged in these behaviors, then maybe we

could save the state consent view with respect to the scope criterion.

There are two main directions this “watering down” strategy might take. The first

maintains that individual states must consent to the individual rules to which they are bound,

but broadens what it takes to give consent. Thus we go from Hobbes’s oath to a signature, to a

handshake, to a nod, to a wink, until we arrive, in Locke’s hands, at tacit consent where mere

compliance or residence within the physical boundaries of an area governed by a legal order

reigns. The second direction suggests that states need not consent to each individual rule, but

to some second order proposition. Hart thought states would only need to consent to a

secondary rule of recognition. A Gilbertian might think that states only need to consent to

becoming part of the collective governed by international legal rules. We could even

85 Hobbes, Leviathan, Book II, Ch. 17, Par. 13.

71

intertwine the two directions to produce a view in which states do only have to accept some

small number of second order propositions and tacit consent suffices.

While these are possible ways to save state consent, I think they are best avoided. The

further we go in these directions, the more strained it is to call what states do “consent,” in the

sense required to bear on questions of rightful authority, especially in view of the ideas of

legal obligation and rights to freedom and autonomy that motivate state consent theory in the

first place. What is worse, even the most watered down version of actual (including tacit)

consent arguably cannot capture all or even most of international law as we find it. As

suggested earlier, even as regards treaty law, it makes little sense to say that states consent to

the principle pacta sunt servanda, the principle by which consent is supposed to bind in the

first instance.

Still following one narrative popular in the western analytic political philosophy

tradition, the next step after tacit consent is hypothetical consent. Rather than ask what states

do accept, we can ask what they would accept, or what they would accept if they were thinking

straight, or behind a veil of ignorance, or willing to cooperate on reasonable or mutually

acceptable terms. But importantly, this is a step that a state consent theorist cannot take. As

Dworkin rightly says, “a hypothetical contract is no contract at all.”86 This does not imply that

hypothetical contract views are to be rejected. But if they are true, it is not because something

is special about consent, as opposed, say, to there being something special about engaging in

particular forms of practical reasoning specially connected to the social forms of life we

participate in.

Gilbert’s work might offer us another way to avoid the scope issues of the state consent

view. She offers a version of the problem of political obligation, what she calls the

86 Dworkin, Taking Rights Seriously, 151.

72

“membership problem”. She writes

It may be that membership in any social group involves obligations. If so, political obligations may constitute the political society version of these obligations of membership. That there are obligations of membership in general is an important part of what I shall argue.87

What she suggests in this passage and subsequently argues for is that membership in a group

may suffice to ground certain forms of obligation. She goes on to argue that these

membership obligations have the form of contractual obligations. Her idea, it seems, is that

contract obligations might actually be a kind of membership obligation, and not vice versa as

is sometimes assumed. Mere group membership, perhaps little more than simple association,

could suffice to ground directed duties. It would be a mistake, however, to call this a “consent”

theory, since membership, not consent, does the normative work.

Whether such an “associative” view could suffice to ground legal obligations in such a

way that it also satisfies the scope criterion depends, this chapter has argued, on how

voluntaristic our conception of group membership is. Certainly agents must explicitly consent

to join many groups like political parties, campus clubs, and marriages, at least in the standard

cases. But we might not consent to joining all groups to which we have obligations. The family

is an important example. It is as certain that we have familial obligations as it is that we do not

consent to who our families are. If we could develop a non-consent based theory of group

membership, perhaps that could suffice to ground the kinds of legal obligation we are looking

for. I will propose one such view in Ch. 4. This might be seen as a version of Gilbert’s joint

commitment account that focuses on the international case and so marginalizes the relevance

of consent as far as possible. In this way, it seems to me that the practice approach is distinct

from her general theory of joint commitment.

87 Gilbert, A Theory of Political Obligation, 18 (italics added).

73

Looking Forward

Consideration of the state consent view illuminates our larger problem of understanding the

justification for contemporary law and the accompanying obligation to conform. In its favor,

the state consent thesis offers an intuitive explanation of the presumed form of international

law. Substantiated by Margaret Gilbert’s account of promising and agreements, this

presumptive form is isomorphic with contractual obligations. However, the voluntarism

essential to the state consent view limits its ability to justify legal duties that defy

interpretation as consensual. Thus the state consent view fails to satisfy the scope criterion.

However, thinking about ways to water down the idea of consent, thus reclaiming larger

portions of the presumed scope of law, suggests promising directions for further inquiry,

though they demand departure from forms of actual consent. The next chapter considers one

such view, presented by a person who is widely considered to be the pre-eminent legal scholar

of the 20th century.

74

Chapter 3: Dworkin’s New Philosophy

Introduction

One of the most promising answers to the doctrinal question of legal obligation in

international affairs is Ronald Dworkin’s posthumously published “New Philosophy” of

international law.88 Dworkin’s own arguments are woven into diverse discussions of history,

policy, and legal theory, but it may be helpful to reconstruct the analytical core of the

argument lying behind Dworkin’s own commentary. I hesitate to lose the richness of the view,

and indeed will draw on a number of additional sources to provide substance where Dworkin

prefers to stay quiet. But my main task in this chapter is to give a clear account of an

encouraging development in the field, as both enhanced by contributions in other disciplines,

and as interrogated with the scrutiny characteristic of modern analytic philosophy.

To oversimplify, the New Philosophy grounds international legal obligation in the

state’s duty to improve its own legitimacy. In the context of Dworkin’s wider body of work,

and along with his accompanying discussion of the “principle of salience”, this claim has led

at least one commentator to take the New Philosophy to clearly be a modern version of

natural law theory.89 This is certainly a natural interpretation of Dworkin’s words, but it is not

the only available interpretation.

In reconstructing the arguments behind Dworkin’s view, I will suggest that we

discover an ambiguity in the New Philosophy. A state’s duty to improve its own legitimacy,

the cornerstone of the view, need not be a natural duty (as Dworkin and at least one

commentator seem to contend). Indeed, Dworkin’s own attention to the sociological

88 Dworkin, “A New Philosophy for International Law.” 89 Compare Lauterpacht, The Function of Law in the International Community; Griffin, On Human Rights.

75

circumstances of international legal practice lends itself to quite diffent, more contextually

sensitive accounts of political obligation. This includes normative accounts that a legal

positivist – Dworkin’s expressed and opposed interlocutor – might adopt. Far from an

objection, the possibility of a more “positivistic” account represents a fruitful avenue for

future research in international legal theory, which I begin to consider in this and the

following chapter.

In this chapter, I take first steps toward outlining a practice-based version of the New

Philosophy, a version grounded, not in natural duty, but in associative obligations. Although it

is essential to his view, Dworkin himself says very little about the relationship between the

legitimacy of a state and the legitimacy of the state system. By clarifying this relation, it

becomes clear that a practice approach is not only a distinctive possibility, but also has the

best chance of surviving philosophical and social scientific scrutiny.

Summary of the New Philosophy

Before offering a sustained analysis of the New Philosophy, I begin with a brief overview of

Dworkin’s basic argument. For the moment eschewing good philosophical practice of

considering alternatives, making clear distinctions, heading off objections, etc., I focus on

briefly summarizing the main outlines of Dworkin’s position, simply to to put a version of the

view on the table, with a promise to provide sustained arguments subsequently.

In our world, political power is divided among semi-autonomous political entities

called states. Each of many states is invested with more or less the same formal right to rule. In

the course of governing well, the state necessarily exerts coercive force on its citizens. Unlike a

highway bandit, coercing passersby at gunpoint, states presume, and are widely presumed by

citizens and other states, to legitimately have a right to exert such force. A state’s legitimacy is

76

what distinguishes state governance, which is permissible, from mere coercion, which is

unjustifiable by presumption. Accordingly, states have a duty to uphold and improve their

own legitimacy, to insure that its exercise of coercion is and remains justifiable to all affected.

Whether or not this has always been the case, as states engage in governance and

exercise the right to rule, they frequently interact with each other in the modern global

context. This has led to a complex social practice of alliances and dispute resolution, trade

relations, and diplomatic standards, which bind states to one another as members of and

participants in a state system. This system is primarily instantiated and constituted by the

states themselves and by the actions and judgments that are recognized by the community of

states. For instance, as we witness the growing recognition of various non-state actors (NGOs),

as well as international organization such as the United Nations (UN), the International Labor

Organization (ILO), and the World Trade Organization (WTO), the international community

of states is increasingly able to (perhaps rightfully) enforce certain rules against particular

members of the larger state system.

Like NGOs, states too must be recognized. Descriptively, states themselves only have

the right to rule in virtue of recognized participation in the state system. The state system,

then, is best understood as a global social practice of mutual recognition.90 Normatively, the

legitimacy of a state’s right to rule depends on the legitimacy of the state system itself. Because

states have an obligation to uphold and improve their own legitimacy, they have an obligation

to uphold and improve the legitimacy of the state system.

But the state system is pathological, which is to say, it normally tends to impose risks of

harm and real injuries on its participants, even in the course of functioning well, according to

90 See, for helpful discussion, Wendt, Social Theory of International Politics, 73–74.

77

its basic underlying purposes.91 Whatever its aspirations, these risks and harms undermine

the system’s legitimacy. If a state is to fulfill its duty to be legitimate, it must take steps to

ensure that the state system itself is legitimate, and so take steps as necessary to address these

pathologies.

International law, then, is the general means by which states, through compliance, can

fulfill this basic duty. That is to say, international law is a body of legally authoritative norms

that addresses the pathologies of the state system, by seeking to ensure that risks and harms it

imposes are reasonably justifiable to those imposed upon, consistent with its basic, politically

decentralized structure. In addressing pathologies by establishing the international rule of

law, the state system avoids being subject to decisive moral objection. By adopting a body of

norms that are presumed to be legally obligatory, independently of their specific content, the

state system can demand obedience despite moral disagreement about individual laws.

Alternative means for upholding and improving the legitimacy of the state system are either

impracticable (like the establishment of a world state), or are not obviously normatively

superior.

Why then should states comply with international law? States have an obligation to

improve their own legitimacy, and so the legitimacy of the state system. The only way to

uphold and improve the legitimacy of the state system is to help address its pathologies, and

the best way to do that is to comply with international law, even if individual laws conflict

91 I borrow the term “pathology” from Aaron James (“A Practice Theory of Global Justice.” unpublished outline), who uses it in a suggestive way in order to distinguish untoward tendencies of a decentralized political system from mere states of affairs or outcomes attributable to other social activities. This rough class has also been termed “standard threats” (Beitz, The Idea of Human Rights.), and “justice sensitive negative externalities” (Kumm, “Constitutionalism and the Cosmopolitan State.”). Although he was one of the first authors to recognize the special normative significance of these harms, and although they are clearly what he has in mind, Dworkin did not provide a clear analysis of them. See, for example, the discussion of coercion on p. 16.

78

with a state’s own conception of morality. States, therefore, have a legal obligation to comply

with international law because, and to the extent that, international law corrects the

pathologies of the state system.

States and the State System

I now turn to a more detailed exposition of Dworkin’s position as he himself presents it. A

useful point of entry is his sociological analysis of the current state of affairs in modern

international law and politics. “The world is divided,” he writes, “into sovereign states, each of

which is in principle immune from interference by other states”.92 Their political powers are

expressed as what we call sovereignty or the right to rule, a bundle of rights and obligations.

These rights include, but are not limited to the right to pass and coercively enforce laws

within a territory, the right to control movement across borders, the right to extract, use, and

sell resources, the right to treaty, the right to borrow money, and the right to forcibly repel

invading forces.93

Whatever may have been the case in the past, or could have been the case in

humankind’s counterfactual departure from earlier history or an imagined state of nature,

today there are many states. Together, these states form a global community called the state

system. Very little is said in Dworkin’s account about the relationship between the state and

the state system, which is surprising since this relationship is essential for Dworkin’s

92 Dworkin, “A New Philosophy for International Law,” 15–16, compare with Krasner, “Sovereignty An Institutional Perspective”; Krasner, Sovereignty. Though he refers to this view as “conventional”, part of the strength of the New Philosophy is that it appropriates central parts of its competitors for its own purposes. 93 Krasner, “Sovereignty An Institutional Perspective”; Krasner, Sovereignty; Lauterpacht, The Function of Law in the International Community; Paul, “Sovereignty, Survival and the Westphalian Blind Alley in International Relations”; Ruggie, Constructing the World Polity.Krasner, Ruggie, Paul, Lauterpacht.

79

argument for the legal obligations of states, and since there is a vast literature, in international

relations theory and political science, about this relationship. 94

This is not to say that Dworkin is silent on the relationship between the state and the

state system, and we can piece together a view from remarks provided. In what I believe is a

key passage, which Dworkin offers as a passing comment, he writes that “each of those states

derives its moral title to govern a particular territory from the arrangements that make up that

international system”.95

To elaborate, Dworkin seems to be suggesting that all states bear the same rights as a

matter of being recognized sovereigns within this “Westphalian” system, which “balkanized”

sovereignty over distinct territories. Yet it would be a mistake to think that states are merely

the passive recipients of commands from the state system, because, after all, it is “the

individual states that make up the system.” What emerges from these descriptions is a picture

of a three-tiered international system. At the top is the Westphalian regime in virtue of which

political power is held by states. States, the second tier, are more or less homogenous entities,

at least so far as the right to rule is concerned (i.e. all states have the same right to undertake

treaties, even if they use it differently; they all have the same right to territorial integrity, even

though they govern over different territories; and so on). States are only sovereign (and only

legitimately so, as we will see) because of their participation in the state system. Individuals,

over whom the state has the right to govern, populate the third tier.

Here it is tempting to conclude that Dworkin is affirming the conceptual priority of

either the state or the state system over the other. After all, if states get their powers from the

94 To name only a few, some of the major contributions to this discussion areChayes and Chayes, The New Sovereignty; Krasner, Sovereignty; Kratochwil, “Limits of Contract, The”; Onuf, International Legal Theory; Ruggie, Constructing the World Polity; Wendt, Social Theory of International Politics. 95 Dworkin, “A New Philosophy for International Law,” 16-19.

80

state system, then the state system must come first, conceptually and perhaps practically

speaking. At the same time, given the current and foreseeable absence of anything resembling

a world state, the state system is instantiated by the interactions between states and their

mutual recognition of one another. So, it may seem, the state must precede the state system.

Here we might compare the “agent-structure” debate, one of the most persistent and

well-explored debates in 20th century international relations theory.96 On one side are the

realists, who argue that the structure of international relations (e.g. the state system)

determines the actions of agents within that structure.97 Structure is explanatorily prior. On

the other side are the liberals, who argue that the actions of agents are what shape the

international system in the first place, and so agents are conceptually prior.98 Constructivists,

notably including Alexander Wendt, try to accommodate both sides of this explanatory

debate.99 State identities and interests are shaped by the world they live in, but as states act,

they shape and reshape the very social world that constitutes and defines them as states in the

first instance.

Dworkin seems to agree most closely with the constructivists. As states act, they

performatively construct the state system. Their actions are performative interpretations and

expressions of the state system itself. At the same time, those actions are constrained,

descriptively and normatively, by the very system being performatively constituted. So

Dworkin might say, with Wendt, that the state and the state system are mutually constitutive.

I return to this point and its importance momentarily.

96 Wendt, “The Agent-Structure Problem in International Relations Theory.” 97 Morgenthau, Politics Among Nations, Revised by Kenneth W. Thompson; Waltz, Realism and International Politics. 98 Kant, Perpetual Peace, and Other Essays on Politics, History, and Morals. 99 Wendt, Social Theory of International Politics.

81

State, System, and Legitimacy

We might further elaborate Dworkin’s basic picture. In the simplest terms, states rule. They

use coercive force to constrain and manage the lives of the individuals who live under them.

But implicit in state rule is the assertion that this rule is not merely coercion, it is governance.

In a recent article, Daniel Bodansky offers a helpful description of governance when he writes

that “Although definitions vary, the essence of governance involves making decisions for a

collective – decisions that not merely affect others indirectly, but are directed at them and are

intended, in some way, to constrain their behavior.”100 Bodansky continues by contrasting

legitimate governance with mere compulsion. Unlike compulsion, legitimate governance

“has a normative quality. A legitimate institution has a right to rule; it is ‘morally justified in

attempting to govern’”.101 By purporting to govern, in Bodansky’s sense, states performatively

assert the legitimacy of their claim to the right to rule. Furthermore, and still as a matter of

description, states and their citizens generally recognize the legitimacy of each state’s claim to

these rights. Though legitimacy can be (and is) contested, states and other international actors

generally respect the claim of states to rule legitimately, according to widely accepted

legitimacy expectations.

At the core of Dworkin’s view about the state’s obligation to comply with international

law is the state’s duty to improve its own legitimacy. Bodansky’s distinction between mere

coercion and governance helps us understand why Dworkin might maintain that states have

such an obligation. State legitimacy is what distinguishes governance from mere coercion, the

state from the highwayman. And although Dworkin is largely silent about the ground of the

state obligation to be legitimate, he clearly affirms it. A natural interpretation is that he means

100 Bodansky, “Legitimacy in International Law and International Relations,” 5. 101 Ibid., 7; quoting from Buchanan, “The Legitimacy of International Law.”

82

to affirm a natural duty of legitimacy, perhaps akin to Rawls’ “natural duty of justice,” which

he takes to be relatively uncontroversial and widely acceptable.

Dworkin does speak to long-standing questions about the normative grounding of a

state’s claim to the right to use coercive force to govern, in the following passage:

The old assumption that hereditary monarchs have an absolute right to govern, at least in the temporal sphere, was gradually replaced by a starkly different assumption: that coercive political power is consistent with the dignity of citizens only if it can be justified not just in pedigree but in substance—in the way it is exercised—as well. Competing theories of legitimacy were constructed and debated; these finally settled into theories about the best conceptions of democracy and of the rights of individual citizens in a democracy. But all these theories were confined to arrangements within sovereign states.102

We might explicate what Dworkin has in mind as follows. In the history of western political

philosophy, there have been two predominant views about the source of state legitimacy. The

first, the “endogenous view,” as we may call it, holds that a state is legitimate if approved or

otherwise authorized by its citizens. The second, the “exogenous view,” holds, by contrast,

that state legitimacy depends on normative considerations outside the state. Though each is

worthy of consideration, Dworkin seems to believe that only the exogenous view can explain

legitimacy in the modern context.

One might think that states become (more) legitimate by receiving approval or other

authorization from their subjects. This is the endogenous view of state legitimacy – legitimacy

comes from within. Such a view would conform nicely to the popular sovereignty tradition.103

But, descriptively, there are many states that appear to be legitimate bearers of the right to

rule, even though they make no pretenses to democracy or popularity. It would seem a

mistake to say that the Saudi Arabian government is not the legitimate ruler in Saudi Arabia,

or the Chinese government in China. These may not be just rulers, given the severe

102 Dworkin, “A New Philosophy for International Law,” 16. 103 Hobbes, Leviathan; Locke, Second Treatise of Government; Rousseau, The Social Contract.

83

oppression and high rates of political and economic inequality in those countries, but they are

certainly the legitimate rulers there.

Moreover, if state power is somehow derived from the will of citizens, one would

expect vast diversity in the form of the state. One nation would have consented to invest its

state with one set of powers, and another nation its state with another set of powers. But this is

not what we see. States may be diverse in terms of the kinds of domestic policies they accept,

but are strikingly homogeneous otherwise. All states have the rights and obligations listed

when we first introduced the idea of the state – the right to govern domestically, control

borders, extract resources, etc. Even newly recognized states, prior to any form of election or

popular demonstration, have these rights. This homogeneity is inconsistent with the popular

sovereignty thesis in particular, and with the endogenous view generally.

Instead, Dworkin encourages us to take seriously the exogenous view that,

descriptively, state power is granted as a matter of global politics. As suggested earlier, in the

beginning, states are members of the state system. But rather than thinking of states as so

many mushrooms, popping up independently in different corners of the world and expanding

until they bump into one another104, we should think of states are political entities endowed

with rights in virtue of their recognized membership in an international society. Put

otherwise, states bear the sovereign right to rule only if and because they are recognized

members of the state system.

“We the People” are not the source of state power, at least not anymore, and we see

this played out in modern practice. Not only are oppressive and unpopular regimes able to

maintain their hold on power by sheer force, these regimes frequently maintain international

recognition as the sovereign governing body in that region. Syria serves as a helpful current

104 Hobbes or even Plato might have held such a view

84

example. This government has killed scores of its own citizens by horrific and brutal means

and has displaced millions more. Nevertheless, this government is the internationally

recognized sovereign of the Syrian territory and population, largely because of the absence of

a widely acceptable alternative governing body. It would be a mistake to say that Assad’s

government is not sovereign in Syria, or that it does not have the internationally recognized

and right to rule. It is and it does.

But Dworkin might have taken this a step further. Not only does Assad’s government

rule, its continued rule is the performative assertion of the right to rule. That is, Assad is

asserting the legitimacy of his government, in both official pronouncements and its failure to

ceded the reins of power to other claimants. By recognizing those assertions, if only by not

publicly contesting them—by letting them stand, as setting the score in global politics—the

international community in effect recognizes the legitimacy of the Assad regime. We can of

course still criticize the Syrian government, and hold it accountable for injustices. But the

deeper assumption is that, for the modern world, state power and legitimacy are not

endogenous – they do not come solely from the will of the people, suitably defined – and so

cannot be justified by reference to internal processes or conditions.105

105 Current events in Crimea also serve as a helpful example. In light of the mid-March 2014 referendum, it appears that the vast majority of Crimeans (even taking into account the Tatars’ and other minority groups’ abstention from the vote) are in favor of secession from Ukraine and inclusion in the Russian Federation. But the international community is far from univocal about the significance of this vote. Russia has recognized the sovereignty of the new Crimean government, even going to far as to cosign a treaty with them. But the United States, the EU, and the Ukrainian government do not recognize the legitimacy of the new government. At the moment I write these words, the following seems to be the right interpretation of the situation: despite the will of the Crimean people, the legitimacy of a sovereign Crimean state is yet to be determined. If the international community comes to recognize the Crimean state, then it will have been a state ever since the referendum. If the international community does not recognize the Crimean state, then is never was a state and has always been part of Ukraine. My point, however, is that it is not the domestic process that will decide whether Crimea is a state. Instead, legitimate statehood depends on in international recognition.

85

Gift-giving might help us better understand the exogenous account of state legitimacy.

When we give a gift, we purport to imbue the recipient with a set of rights to that gift, usually

the rights of property. But for these rights to be genuine, for the receiver to be the legitimate

owner of the gift, the givers must have been in a normative position to imbue those rights. In

the usual instance, gift-giving is a transfer of ownership. If I am the owner, I am in a position

to transfer my rights of ownership to you. If I am not the owner, I may be able to physically

hand an object to you in a way that resembles gift-giving, but I cannot imbue you with the

rights of ownership.

Something similar may happen between the state and the state system. The right to

rule is like a gift. When a state (or proto-state) is recognized as sovereign by the state system,

the state system purports to imbue the state with a new set of rights. These rights are the right

to rule. But in order for this allocation of rights to have genuinely occurred, in order for the

state to be the legitimate bearer of the right to rule, the state system must have been in a

normative position to give those rights in the first place. To use Dworkin’s terminology, in

order for the state to be legitimate, the state system must also be legitimate.

So, on this view, in order for a state to satisfy its obligation to be legitimate, two

conditions must be met: First, the state must be a recognized participant in the state system;

second, the state system must be legitimate. Dworkin is very clear on this point when he

writes:

It follows that the general obligation of each state to improve its political legitimacy includes an obligation to try to improve the overall international system. If a state can help to facilitate an international order in a way that would improve the legitimacy of its own coercive government, then it has a political obligation to do what it can in that direction.106

106 Dworkin, “A New Philosophy for International Law,” 17.

86

To highlight the point, Dworkin adds, “That requirement,” what he calls the “duty of

mitigation”, “sets out, in my view, the true moral basis of international law.”107

The Legitimacy of the State System and the Duty of Mitigation

The question to answer, then, seems to be: What must states do to uphold and improve the

legitimacy of the state system? But before this question can be answered, Dworkin thinks that

we need a sense of what could threaten the legitimacy of the state system in the first place.

Just as a physician must diagnose the cause of a pain before she can prescribe a cure, so we

must identify the potential sources of illegitimacy in the state system, before we can tell sates

how they ought to address them. With this intention, Dworkin offers a discussion of “the

different ways in which individual states fail their responsibilities to their own citizens when

they collectively accept the benefits and burdens of the pure unrestricted sovereignty that the

Westphalian system gives them.”108 He elaborates as follows:

These are all ways in which the unchecked state sovereignty system impairs or threatens the legitimacy of the individual states that make up the system. But since each of those states derives its moral title to govern a particular territory from the arrangements that make up that international system, it therefore has the further, independent reason…for concern that the system on which its legitimacy depends in that more fundamental way is not itself illegitimate.109

What Dworkin proposes in these remarks is that there is a specific class of morally salient

problems, specific to the “unchecked state sovereignty system”, that threaten the legitimacy of

that system and so the legitimacy of states.

107 Ibid. 108 Ibid. 109 Ibid., 19.

87

What are these problems, and how do they damage the legitimacy of the state system?

Dworkin himself offers very little guidance on this question, except that he gives four helpful

examples of the kinds of problems he has in mind. To paraphrase, they are as follows:

1) Because states in the modern Westphalian state system are allowed a great amount of

autonomy in deciding the structure of domestic governance, citizens of states are

exposed to the risk of tyranny and oppression by their own government.

2) Because states in the modern Westphalian system are permitted to autonomously

govern themselves, these permissions serve as institutional barriers to justified

intervention and aid when people in foreign countries are in need.

3) Because political power is decentralized in the Westphalian system, otherwise

autonomous states must coordinate with one another to address cross-border issues.

But such collective action can be difficult in prisoner’s dilemma and stag-hunt type

situations.

4) Because states in the modern Westphalian system are allowed autonomy in

determining the structure of domestic politics, citizens are at risk of not being afforded

their right to political representation.

Certainly these are not the only adverse tendencies of the state system, though these suggest

paradigmatic examples. We might further characterize the larger class of problems at issue as

follows.

First off, the “problems” in question are risks of harm and real injuries that agents are

forced to bear while living under the state system. Generally, however, they are attributable

to the state system, in virtue of untoward tendencies or “pathologies” that arise from its

politically decentralized nature.

88

This is to exclude hardships due simply or mainly to bad luck, which may not be

caused by human activity, such as harsh weather that washes away a farmer’s crops, a person

or electrical grid being struck by lightning, or a sudden state-wide water shortage in desert

countries. While many such events may be addressed through precautionary social policy, to

the extent they cannot be prevented or entirely rectified, the resulting hardships would not

count against the state system’s legitimacy. Not only are natural accidents and acts of God

problems for any political order, they cannot be causally linked, however indirectly, to the

state system itself. These kinds of problems, therefore, do not ground reasonable objections to

the legitimacy of the state system.

The relevant class of problems also excludes mere “bad apple” actors, who abuse or

otherwise misuse power, and the costs they impose upon others within the larger system.

These costs and risks may be unjust, but to the extent they are not inherent problems with the

state system itself, but to actors working within its structures, they may not reflect illegitimacy

in the larger system. Example may include certain tyrannical leaders and rogue states, which

knowingly violate the obligations of responsible statecraft; the inadvertent deaths during the

2013-14 Maidan protests in Ukraine; or the misappropriation of funds during Berlusconi’s

administration. In such case, there is at least a case to make that the injustice is not

sufficiently attributable to the structure of the state system, as opposed to objectionable

individual decisions that agents make within that system (though of course one can equally

argue that a more legitimate system would adopt practices that make such abuses less

common that they otherwise might be).

With this fuller understanding of state system “pathologies” in hand, we are in a better

position not only to understand Dworkin’s examples, but enumerate further untoward

89

consequences of decentralized global governance in the post-war and modern eras. These

include:

Risk of Domestic Tyranny: Because states have the right to pass and enforce laws internally, and

other states are prohibited from interfering, individuals are left unprotected from

domestic inequality, oppression, persecution, ethnic cleansing, genocide, and any

number of other horrible things a government can do to its own citizens. Individuals

are not protected against the tyranny of their own state.

Environmental Coordination Problems: Because states are granted exclusive control of a territory,

many of the environmental problems with which we are currently faced demand

coordination across borders. However, such large scale and costly cooperation is hard-

won, slowly adopted, and easily undermined in a system of sovereign states at the

potential risk of global ruin. Importantly, this is not simply the product of a few bad

apples – a selfishly obdurate developed state or an insatiable developing state with an

appetite for coal, say – but a consequence of any attempt to get any group of complex

but otherwise autonomous bodies to do the same thing in a coordinated way at

personal cost.

Economic Coordination Problems: Similarly, as domestic economies are increasingly exposed to

international markets, maintained economic stability has become a matter of

coordinating domestic policy. Because economic decision-making and governance is

decentralized, states are effectively stuck with the problem of other minds and the

need to coordinate with those others nevertheless.

Risk of Unfair Distribution: Because states control fiscal and economic policies, and because

different policy regimes can have wildly different consequences, the state system is

consistent with severe socio-economic inequality between states.

90

Reduced Security Assurance: Because states are mostly independent of one another, they are not

always, or ever, certain about one another’s military capabilities and ambitions. The

system provides no assurance, in whatever form, that states will not aggress against

one another. A state is only as safe as its own diplomats and military can keep it, and

this uncertainty can lead to mutually undesired escalation and vast overinvestment in

security.

Exploitation Risk: Similarly, the state system does not protect weak and poor states from

exploitation by strong and wealthy states. A small state, whose citizens’ livelihoods

depend on trade with a stronger state, is at the mercy of the stronger for fair terms and

compliance.

Lack of Representation: Finally, by granting states the right to exclude external governance

regimes, the state system reduces the political power of many people, now categorized

as “non-citizens” or “aliens”, whose lives are deeply effected by a country’s policies.

American presidential elections, for example, have vast implications for the lives of

individuals around the world, but the American president is only elected by and so

accountable to the American citizenry.

In such cases, I suggest, certain morally salient risks and harms are evident, soluble by

available means, and emerge as byproducts of the state system, even when it working as it is

supposed to according to its basic purposes. In theory at least, such tendencies might arise

under conditions of impeccable luck and perfect compliance.110

110 In conversation, Nicholas Onuf once tried to sum up my conception of pathology as a sort of dysfunction, when, to paraphrase his words, “institutional constraints make the mutual satisfaction of interests impossible.” He was correct to invoke the idea of function, but the matter, I think, is more complicated than he made it seem. First, it seems to me to be a mistake to think that the function of the state system is the mutual satisfaction of state interests. As we have seen, I take the function of the state system to be the realization of more specific values (peace, prosperity, and pluralism). Second, I am hesitant to adopt the language of dysfunction,

91

The larger idea, then, is that the state system can only be regarded as legitimate when

such pathologies are being addressed. Only then can states justifiably claim that their

exercise of power is more than mere coercion, and is true governance. In Dworkin’s terms, a

state’s obligation to improve its own legitimacy implies an obligation to improve the

legitimacy of the state system, and that, in turn, implies an obligation to address or “mitigate”

its adverse tendencies.111

International Law, Compliance, and the Principle of Salience

It remains to be seen how this duty translates into a duty to comply with international law. In

this connection, Dworkin writes,

That duty of mitigation provides the most general structural principle and interpretive background of international law. But as it stands, it is not sufficiently determinative. In many circumstances, a number of very different regimes of international law would each serve to improve the legitimacy of the international system, were it enacted and enforced, and states may reasonably disagree about which would be best.112

So even with the duty of mitigation in hand, we are stuck with a puzzle about the obligation to

comply with the law. Dworkin takes for granted that some legal system or other will be the

best or perhaps only way for states to address the pathologies of the state system. But even

granting that thought, Dworkin worries that multiple versions of a given international legal

system will be equally well suited to the task. Given that international law can be informally

legislated, as in the cases of customary and administrative law, an account of this cannot

which, to my mind, invokes the idea of a system that breaks down. A dysfunctional car does not run. But the idea of pathology is not so dramatic, and includes references to optimality as well as mere function. A pathological system may be one that merely imposes unnecessary risks, without being wholly dysfunctional. This is one of the important points about pathologies: they count as meaningful objections to a practice because they can be the products of otherwise perfect functioning. 111 Dworkin, “A New Philosophy for International Law,” 19. 112 Ibid.

92

simply invoke a procedural secondary rule of recognition. But how then are we to know

which version of international law is the law?

Dworkin’s answer is what he calls the “Principle of Salience”:

If a significant number of states, encompassing a significant population, has developed an agreed code of practice, either by treaty or by other form of coordination, then other states have at least a prima facie duty to subscribe to that practice as well, with the important proviso that this duty holds only if a more general practice to that effect, expanded in that way, would improve the legitimacy of the subscribing state and the international order as a whole.113

The basic idea of salience is that if, within a population or society, there is an identifiable and

predominant set of rules governing social interactions, participants have a prima facie duty to

follow those rules as well. This is more than the platitude, “When in Rome, do as the

Roman’s do.” Dworkin’s “proviso” conjoins the socially relativistic idea of salience and the

normatively rich principle of mitigation. States have a prima facie duty to comply with going

international legal practice, but only really have such a duty if that legal practice in fact works

to mitigate pathologies of the state system.

Natural Law and the New Philosophy

In sum, then, for Dworkin, a state’s duty to comply with the law stems from its duty of

mitigation, which itself stems from the duty to be legitimate. But it is not obvious, on its face,

what the basis of this duty is supposed to be. Since the initial circulation of the new

philosophy (at least c. 2010), perhaps the most helpful contribution to the associated literature

has been Matthias Kumm’s essay “Constitutionalism and the Cosmopolitan State”. Kumm

offers us an account of not just one but two different justifications for the duty to be

legitimate. Very much in the same vein as Dworkin, Kumm writes:

113 Ibid.

93

One of the core purposes of international law is to create and define the conditions under which a sovereign state’s claim to legitimate authority is justified. States have a standing duty to help create and sustain such conditions and an international legal system that is equipped to fulfill that function.114

But Kumm immediately distinguishes his own view from Dworkin’s in an attached footnote

when he writes:

But whereas Dworkin’s account appears to be focused on the ways in which sovereigns are structurally incapable of adequately securing the rights and public goods for their respective citizens, the argument here grounds cosmopolitan obligations in the fulfillment of duties of justice towards outsiders.115

According to Kumm, Dworkin ought to be interpreted as saying that states have an

obligation to improve their own legitimacy because only by doing so will individual persons

be afforded their basic rights and needs. This seems like a fair interpretation, given Dworkin’s

explicit concern with the treatment of individuals by their governments in statements like the

following:

A coercive government is of course illegitimate if it violates the basic human rights of its own citizens. Any state, even one that has so far been just and benign, therefore improves its legitimacy when it promotes an effective international order that would prevent its own possible future degradation into tyranny.116

On this reading, individuals are the bearers of a set of natural rights, largely akin to the list we

find in the UDHR or the American Bill of Rights. These rights ground correlative duties for

the state. Among these duties, or perhaps taking these duties as a whole, is the state duty to be

legitimate.

114 Kumm, “Constitutionalism and the Cosmopolitan State,” 8. 115 Kumm, “Constitutionalism and the Cosmopolitan State,” 8n12. 116 Dworkin, “A New Philosophy for International Law,” 17. The following quote is also helpful: “People around the world believe they have—and they do have—a moral responsibility to help to protect people in other nations from war crimes, genocide, and other violations of human rights. Their government falls short of its duty to help them acquit their moral responsibilities when it accedes to definitions of sovereignty that prevent it from intervening to prevent such crimes or to ameliorate their disastrous effects.” (17-18).

94

Kumm himself adopts a different, but closely related view about the duty to be

legitimate, though it, too, is ultimately grounded in natural rights. Kumm agrees that

pathologies – what he calls “justice sensitive negative externalities” – are indeed the target of

international law’s corrective function. But unlike Dworkin, Kumm argues that states have an

obligation to comply with international law to solve these problems, rather than solve them

independently, because

Given the fact of reasonable disagreement between states about how those externalities should be taken into account, any claim by one state to be able to resolve these issues authoritatively and unilaterally amounts to a form of domination.117

Rather than the human rights enjoyed by individual persons, Kumm grounds the state’s

obligation to comply with international law in other states’ rights not to be dominated. Kumm

offers little argumentation in favor of his claim that states have a standing right against

domination, apparently taking the view to be self-evident, or plausible enough as a point of

departure. For example, he writes:

An imperial policy of domination and expansion subverting the political and territorial independence of neighbors is obviously not justified, even when such a policy enjoys widespread democratic support in the aggressor state and that state has a well-structured national constitutional system.118

Kumm’s point is intuitively attractive; there does indeed seem to be something wrong

with forms of international domination. One of the great virtues of the modern state system is

its institutionalized respect for pluralism and diversity. Political power, as we said at the very

beginning and as Kumm emphasizes himself, is decentralized. It is distributed in the form of

the right to rule to semi-autonomous political bodies associated with a territory and its people,

and these bodies are allowed to exercise that right as they will. And as we see, this

decentralization has produced variety, even if the scope of that variety is bounded by 117 Kumm, “Constitutionalism and the Cosmopolitan State,” 9. 118 Ibid., 18.

95

international law. Domination is the antithesis of this pluralism and diversity, so we can

understand where Kumm is coming from.

More important for present purposes, Kumm’s analysis provides us with a second

version of the new philosophy, and one that is strictly international. Setting aside whatever

rights citizens may bear, states themselves are the bearers of the right not to be dominated.

This right, like Dworkin’s human rights, correlates with a state’s duty of mitigation, which in

turn grounds the duty to comply with international law.

Yet the problem, or at least one problem, with both of these interpretations of the new

philosophy is that neither plainly grounds an argument for the claim that international law

has the characteristics of authority presumed in practice. Specifically, neither the human

rights interpretation nor the non-domination interpretation can justify the claim that states

have a duty to comply with legal rules even when they contradict moral truths. I therefore turn

to a fresh approach.

An Associative Obligation to Mitigate

There is another way to interpret Dworkin’s New Philosophy, one that is friendlier to the

positivist in ways that make it better able to capture the characteristics of authority implicit in

practice while also being (more) consistent with Dworkin’s broader contributions to legal

scholarship. We have assumed so far, following Kumm, that Dworkin’s account of legal

authority and the correlative obligation that states have to comply with law is ultimately

based in a natural law conception of human rights. On such a view, humans are the bearers of

certain moral protections that give others correlative duties including, importantly, the duty

to mitigate. This dependence on moral truths like natural rights and duties is what gets

Dworkin and Kumm in trouble with content-independence.

96

But in Law’s Empire, perhaps his most definitive treatment of the normative case for

compliance with the law, Dworkin offers an account of associative obligation as a ground for

the authority of law. He describes associative obligations as

the special responsibilities social practice attaches to membership in some biological or social group, like the responsibilities of family or friends or neighbors. Most people think that they have associative obligations just by belonging to groups defined by social practice, which is not necessarily a matter of choice or consent, but that they can lose these obligations if other members of the group do not extend them the benefits of belonging to the group. These common assumptions about associative responsibilities suggest that political obligation might be counted among them.119

The idea, it seems, is that there is a kind of obligation that agents can bear simply

because they are members of a social group. For example, as a brother, I owe a certain amount

of care and respect to my sister that I do not owe to strangers because of the familial relation

that my sister and I share. The example of obligations between siblings is, I think, especially

useful because they highlight the non-voluntariness of associative obligations. But it is

important also to stress that the family and other so-called natural associations are not the

only kinds that create associative obligations. We have obligations to our friends, our

neighbors, and the parents of our children’s classmates. And none of these are necessarily

because we have consented to them (you can only choose your neighbors when you are the

one moving) or because we have natural duties to be good neighbors (forcing people who live

in rural areas to drive great lengths to lend a cup of sugar).120 Rather, we bear these

obligations because we find ourselves already related to others in these relevant ways, and

that relation alone suffices to bind us to certain standards of conduct. So instead of

interpreting Dworkin’s new philosophy as based in a doctrine either of human rights or of

119 Dworkin, Law’s Empire, 195–196. 120 Dworkin explicitly presents his account of associative obligations as an alternative both to tacit consent and to natural duties.

97

non-domination, we might think of it as an account of the associative obligations that states

bear to one another on the basis of their common membership in a global community.

Not only is this interpretation of the new philosophy consistent with Dworkin’s other

writings, it resonates with his remarks within the posthumous paper as well. He insists that

states are part of a state system, and specifies its balkanized and decentralized character. He

emphasizes a certain set of problems, those produced by the kinds of relations that exist

between states, as particularly harmful to legitimacy and therefore relevant to questions of

international law. He tries to incorporate the beliefs and judgments of practitioners, the

traditions and practices that the subjects of the law share. These details are difficult to place

within a natural law style derivation of states’ duties from human rights or from rights against

domination, but they make sense within an account of associative obligations.

Here, then, is a sketch of the New Philosophy as grounded in associative obligations.

In our world, sovereignty is an institutionally defined right to rule. Domestic protective

associations (i.e. pre-states, or nominal states, or state-like-things-prior-to-recognition) gain

the sovereign right to rule only by being recognized as a state by other states. Together, these

states populate the state system, a social practice that both distributes and legislates the terms

of rule.

This system is presumed to be good for many reasons but importantly because it

promotes shared and widely acceptable goals including human dignity, prosperity, and

security. But by pursuing these goals through such a decentralized system of governance, the

system creates harmful side effects. Two basic mechanisms have emerged within the state

system to address these pathologies. First, the state system has developed a number of

mechanisms for creating norms general compliance with which would help mitigate the state

system’s untoward tendencies. Among these mechanisms are the sources of law specified in

98

Article 38 of the UN Charter - treaties, custom, and so on. Second, the state system has -

through these mechanisms - developed a number of regulatory norms presumed to bind

states such that, absent special countervailing considerations, they ought to comply with those

norms regardless of what they command.

When states claim the right to rule - the rights of sovereignty institutionally specified

by the state system - they become members of this flawed but presumptively justifiable

practice, and as members, undertake an obligation to the other members to do their part in

mitigating pathologies. And in the current world context, that obligation can only be

discharged by complying with international law.

I now turn, in the final chapter of this dissertation, to develop the foregoing practice

approach in greater detail. We will see that it not only accounts for the presumed authority of

international law in a “positivistic” way, but also that it has considerable attractions.

99

CHAPTER 4: The Practice Approach

Introduction

I have suggested that Dworkin’s New Philosophy offers at least two important insights

into the question of why states would have reason to comply with international rules. The first

insight is about the functional role of international law within the state system, which may be

called the

CORRECTION THESIS: international law as we find it, including treaty,

administrative and customary law, is in fact a remedy for problematic tendencies of a

politically decentralized state system (e.g. legal norms that grant diplomatic immunity

help address coordination problems and miscommunications which can precipitate

confusion and conflict; environmental law such as the Kyoto Protocol coordinates

multi-state action in response to global climate change).

The second insight addresses what significance this might have for state conduct. It may be

called the

DUTY OF MITIGATION: absent special justification, states have sufficient reason to

adopt policies that help to remedy problematic tendencies of a politically

decentralized system.

So to the extent that state compliance with established international rules in fact corrects for

such tendencies, states have sufficient reason to comply with treaty, administrative, or

customary law as it may apply to them, at least absent some special justification that excuses

them from performance.

If my account of the authority of international law in Ch. 1 is correct, this is not yet an

account of why international law is authoritative for states. While it does address the “doctrinal

100

question” of why states have reason to obey, to the extent the Duty of Mitigation is seen

simply as a natural duty — as it is for both Dworkin and Kumm — reasons for state

compliance with international law will be too closely linked to the morality of its content. As

proposed in Ch. 1, going practice assumes that law can have a form of legal authority that,

crucially, persists despite inconsistency with widely accepted moral truths. Positivist views

that appeal to state consent accommodate this concern about the form of authority, but, as

argued in Ch. 2, in a way that is undesirably revisionistic. In this chapter, I outline a positivistic

version of Dworkin’s New Philosophy that better captures both the form and scope of going

international legal practice than any view so far considered. Following the work of Aaron

James, I call this the practice approach to international law.

According to the practice approach, states have an associative obligation to comply

with international law. Whatever further natural duties states may or may not have, they have

sufficient moral reason for compliance, absent special justification, by virtue of the specific

social relationship that defines them as states in the first place. By participating and being

recognized by others in the state system, states claim the rights of sovereignty but also

undertake an associative obligation to mitigate untoward tendencies arising from its

politically decentralized nature.

In addition to the Correction Thesis, the practice approach may be expressed as the

following further three claims:

SOVEREIGNTY THESIS: sovereignty, or the right to rule, is not a function of domestic

policy or the choices of peoples, independently of international relations, but rather a

bundle of rights to rule that are institutionally specified through the complex social

interactions of states and other international political agents. The international system

101

not only specifies the moral right to rule over a territory, it distributes those rights

through mutual recognition, of and by states.

ASSOCIATIONAL DUTY OF MITIGATION: absent special justification, states have

sufficient reason to adopt policies that help to remedy problematic tendencies of a

politically decentralized state system, simply by virtue of presumed membership in the

state system, which is to say, by virtue of credibly claiming the right to rule over a

territory with the privileges the state system affords.

AUTHORITY THESIS: the associative obligations of states have the following

authority characteristics: (1) they give states directed reasons for action, such that

compliant actions are owed to other states in a way that is distinct from advice and

requests (“reason-giving”); (2) the obligation to comply with established international

legal rules is defeasible, provided sufficiently important countervailing concerns

(“defeasibility”); (3) the content of such legal rules is set in part by going state practice,

and so is not a direct function of the morality of its content (“content-independence”).

My claim is that these three theses suffice to account for the authority of international law as

we find it in contemporary practice. Although the account is “positivistic” in some ways, it is

unlike state consent views in that it applies broadly across treaty, administrative, and

customary law, and so need not be excessively revisionistic.

My discussion of this account will focus on motivating and elaborating the foregoing

main theses. I address objections, but do not claim to have defended a practice approach

102

against rival views. My main aim is to articulate the general contours of a one way a practice

approach might go.121

A General Account of Associative Obligations

In most general terms, associative obligations are standards for right conduct that bind agents

on the basis of their membership in a group or participation in a social practice. Such

obligations are very common. They include neighborly obligations to lend your neighbor a

cup of sugar, filial obligations to treat a sibling with respect, and, more controversially,

political obligations to pay one’s taxes. But what is it about these relations that suffices to

ground obligations?

Consider siblings. Brothers and sisters are generally thought to have rights and

obligations against one another, which are based in their relationship. For instance, if my

sister were to invite me to her wedding, I would have an obligation to attend, absent

compelling extenuating circumstances. Of course, anyone who is invited is allowed to come,

but not necessarily required to attend, on pain of being open to special forms of accountability

seeking, including public scorn, shame, and resentment. If I were to skip the wedding, but

then complain of my sister’s and my family’s scornful reaction, protesting “Why punish me

so?,” or even “Why punish me and not others who also missed the wedding?,” her answer

would almost certainly be: “I thought you were my brother!” Some views might take her

answer to be elliptical for a longer answer about the rights of human beings or the sanctity of

tradition. A theory of associative obligation treats them in a straightforward way, by taking

121 Another way one might interpret this chapter is as an attempt to give what she has referred

to in conversation as an “operational definition” for non-consensual versions of Gilbert’s joint commitments.

103

reference to the relationship between individuals as the basis for justifying this class

normative claims.

The idea of associative obligations may also be applied, albeit in a more controversial

way, to the practice of promising. Suppose that I promise to paint the house, and you accept

my promise. Now suppose that, for no particular reason, I fail to perform. My failure puts you

in a position to appropriately rebuke me or otherwise hold me accountable. Maybe I owe you

an apology or even compensation. But why? Perhaps it is because there is a true moral

principle, pacta sunt servanda, and the truth of that principle implies the wrongness of my

action and the justifiability of your rebuke. Perhaps it is because I consented, thereby

wrapping myself up in a contradiction when I went against my own will and did not paint the

house. Perhaps it is because everyone, including me, benefits by having a working institution

of promising, and keeping my promises is how I pay my fair share. At the same time, we can

also take the answer to be as simple as what we would actually say: “Because you promised!”

My invoking the practice of promising, that is, by having made the promise in the first place,

might suffice to explain why I bear obligations of that practice.

This raises the question of when association with other agents obliges one. The

foregoing examples - familial obligation and promissory obligation - might seem to have very

little in common. People normally choose whether or not to make promises, whereas hardly

anybody has a say in whether they will have siblings or not. While promises last only until

they are discharged, familial obligations are ongoing, and cannot be fully discharged at any

given time, short of a significant disruption of normal relations. Worse, many social

relationships do not generate obligations at all. Living in the house next to yours may give me

neighborly obligations, but the fact that I work in the skyscraper next to the one you work in

doesn’t, or at least this is much less obviously true.

104

What is obvious, or at least what I believe is apparent in this diversity, is that it might

be too much to ask for a single set of necessary and sufficient criteria for when a relationship

grounds associative obligations. I will therefore simply develop an extended analogy. By

considering how associative obligations might arise in a fictional club, we can then see, by

analogy, how a similar story might apply to international political practice. (I beg the reader’s

patience with an extended presentation of a fictional situation and what will be crucial

details.)

A student group on campus is a pretty cool club - well known, respected, with some

level of prestige and honor – and so the club is exclusive to some extent. Members are given

special privileges, with regular social offerings and connection to a vast network of club

alumni after graduation, both of which give access to exclusive opportunities. As members

graduate, new members are invited to join, and with a continual flow of new and departing

members, leadership is in a state of constant flux with dominant members or alliances coming

and going.

But there are problems. Members disagree about who should be admitted. One

member wants the group to be more inclusive, to share the bounties of membership and to

grow the alumni pool. After all, he says, the point of this whole group is to meet new people

and make new connections. Another member disagrees, preferring to be more exclusive to

improve the quality of the alumni pool and the character of the people with whom she will

associate herself. After all, she says, the point of the association is personal refinement and

more secure job prospects. The quarrels about who should be kept are more fraught than the

discussions about who should be accepted, especially with persistent grumblings about the

slackers whose grades have fallen off or who have switched their majors from pre-med to

philosophy, who therefore don’t hold promise for improving the alumni pool for future

105

members. The slackers, it is feared, may deter promising applicants from joining the club even

once they are accepted, and so, it is proposed, it may be for the best for their own privileges of

membership to be revoked.

The problems are not simply disagreements. While all can agree that it makes no sense

for every member to be in touch with every alumnus to keep the alumni network alive—if

only because the redundancies waste time—no member prefers to work alone, or even in a

small group, for fear that the others will enjoy the benefits of their labor without doing their

part. So the more ready contributors devise creative ways to distribute the workload, making

sure that everyone does his or her part. Here it could well have happened that, with so many

competing options and a lack of sufficient support for any one, the alumni network falls apart

as the club simply loses touch. But fortunately, the vision of a particular alliance of members

did, in earlier years, manage to take hold. Now, in accord with the practice established, the

members divide up the alumni contact lists more or less evenly, and each member stays in

contact with only those few people. The lists are redistributed every year to account for the

changing membership. They begin accepting only people who are willing to do this, but who

also meet standards of promise with respect to the improvement of the alumni network (they

have high GPAs, are ambitious, are independently well connected). They also develop

standards for membership: minimum GPAs, regular participation in group activities, a good

record of staying in contact with assigned alumni.

To add one final elaboration, let us also imagine that these standards change over

time, sometimes subtly and sometimes abruptly. At certain times, the club weighs GPA

heavily in admissions decisions, while, at other times, it seeks applicants with ambitious

majors. The same is true for membership standards. When the club is doing well because

applicant interest is high and the alumni are successful and closely connected, standards for

106

keeping members might rise, in view of greater demand. When the club is doing poorly, and

the alumni are lethargic and unwilling to donate money, standards for acceptance might drop

for new entrants, even while standards for continued membership might rise. In some cases,

the changes are abrupt, as during the year one member called attention to the fact that what

was originally a co-ed club has become dominated by men, largely due to arbitrary gender

biases implicit in the application process. Not only is this exclusion immoral, the member

argues, it is killing the quality of the alumni network, which, after all, is one of the main

reasons for the club. While, initially, one outspoken person had charismatically called for

demographic quotas, while loudly shaming all who failed to comply, the practice caught on

over time, mitigating the problem of gender imbalance. Eventually, the problem was largely

resolved.

We can now pose the question of obligation: Do the members have an obligation to

comply with the rules of the club? Do they owe it to the other members to abide by the

demographic acceptance quotas, which might themselves raise moral questions, when

evaluating applications? Are they rightly sanctioned, or even expelled, by the membership

when they fail to do so? Or when they fail to keep in touch with their assigned alumni, even

though they cannot choose which or how many alumni they are assigned? The answer would

seem to be yes. The members owe it to one another to contact their alumni, and to keep their

GPAs up, and even to use the demographic quotas while evaluating applications. Why?

Because, and simply because, they are members. Or so one wants to say.122

As for why membership should oblige, we might elaborate by suggesting general

conditions that appear jointly sufficient to ground associative obligations, as follows.

122 As mentioned in Ch. 2, Margaret Gilbert discusses membership as the foundation for political obligations. The

practice account might be helpfully thought of as the development of an “operational definition” for such membership commitments which, to my knowledge, has not been previously done. See Gilbert, A Theory of Political Obligation.

107

(1) For starters, the club is a social practice - a group of agents whose behaviors can be

understood as individual activity coordinated by a set of norms.

(2) These norms are organized around generally accepted but appropriately vague

purposes (e.g., as explained in terms of “full social calendars,” “job prospects,” etc.). Such

norms have two basic functions: regulative and constitutive. On the regulative side, certain

norms set the standards for member behavior, and are presumed by members to ground

accountability seeking behaviors when they are violated. On the constitutive side, certain

basic norms define the identities of the agents. If I am a member of the club, I am someone

who keeps such and such GPA, who uses such and such admission guidelines, who calls these

assigned alumni, who attends these events. (Many if not all of these norms may have both

regulative and constitutive roles.)

(3) Adjustments to these norms also fulfill a corrective function. For reasons outside of

the club’s control, leadership is persistently fragmented and in flux, and this leads to

problems that most everyone recognizes. Not only do morally objectionable practices pop up,

the club begins to fail to effectively pursue its own purposes, even when everyone is willing to

do their part. The norms about admission, membership, and networking coordinate the

activities of the membership and correct these perceived problems.

(4) Such norms provide appropriate grounds for criticism and sanction among

members, simply in virtue of membership. Even if it is important that enough members act as

though they owe it to the others to comply, in ways that offer general assurances of

participation, and it needn’t be true that every member has consented to every norm in this

organization for sanction to be appropriate. Nor is it the case that every member has

consented to some decision procedure for determining rules. Nor must it be true that every

constitutive rule of the club is consistent with morality. The club needn’t be rotten, and may

108

not be all that bad, for helping people get jobs after graduation and cultivating a sense of

belonging. But it just as certainly is the case that if a member fails to maintain the minimum

GPA, even if that standard has changed since his joining, that he is liable for sanction or

expulsion. And it is the case that if he does not use the customary standards for admissions, he

is liable for sanction as well.

Why should this be so? Because, I suggest, when one claims the rights of membership

– the prestige on campus, permission to attend exclusive events, access to the alumni network

– one at the same time undertakes an obligation to uphold and improve the legitimacy of the

group or social practice that grants those rights. This view shares much in common with the

fair play argument, most closely associated with Rawls and Hart.123 According to this

argument, roughly, when one receives the benefits of a system of social cooperation, one

incurs an obligation to bear the burdens that make those benefits possible. Or in John

Simmons’s version, one becomes obliged when one “accepts” a system’s benefits, even

without a deliberate, knowing undertaking of an obligation. As Simmons explains:

No deliberate undertaking is necessary to become obligated under the principle of fair play. One can become bound without trying to and without knowing that one is performing an act which generates an obligation. Since mere acceptance of benefits within the right context generates the obligation, one who accepts benefits within the right context can become bound unknowingly.124

The argument that I am advancing is distinct from the fair play argument in at least

three immediately relevant respects. First, the obligations member states incur are not

conditions for playing fair, so as not to exploit the cooperation of others, but for legitimate

rule. When political groups claim the right to rule, they enjoy that right only on terms

specified by the international system, because the international system itself sets the

123 Hart, “Are There Any Natural Rights?,” 55; Rawls, “Legal Obligation and the Duty of Fair

Play.” 124 Simmons, Moral Principles and Political Obligations, 116–117.

109

conditions for the legitimate exercise of political power. Second, it is not necessary that agents

accept benefits form a scheme of social cooperation to be bound. Rather, I am proposing that

agents must merely claim a set of rights, whether or not these rights actually result in any

relative or absolute gains. If an agent does benefit, what matters is not its possession of those

gains, but at least its claim that those gains are its own.125 This explains why an agent (such as a

state or political community) would continue to be bound by the rules of a system of

cooperation even in the presence of evidence that he or she would have done no worse (but no

better) without that system. Second, the resulting obligations are not to do one’s part in

bearing the burdens that make cooperative benefits possible. Rather, they are to comply with

the established terms and to create new terms that will further advance the legitimacy of the

social system.

Here one may ask: Why wouldn’t one be obliged simply to comply with some basic

stated and previously established terms? Because, I suggest, a member’s claim to the

institutionally specified rights and privileges granted to members is only as strong as the

practice’s right to distribute or withhold them. And so to the extent the practice has moral

problems that threaten its legitimacy, this claim to its rights and privileges is not secure. Such

problems open the practice to moral objections that, if not duly resolved, threaten to turn the

practice from a way of socially pursuing widely acceptable goals into a way of sustaining

arbitrary inequality and discrimination. The best understanding of the practice, that is, may

call its animating purpose into question, and so call its rights and privileges into question as

well.

How can a member satisfy the obligations he or she incurs? In several ways. One is to

defer to existing corrective norms where they exist and fulfill the corrective function. Another 125 Incidentally, this subtle alteration might offer grounds to reject Nozick’s objections to the

fair play argument raised in Nozick, Anarchy, State, and Utopia, 93–95.

110

is to amend existing norms in whatever legislative capacity members are granted by the

practice, so that they better correct for its untoward tendencies. Still another is to help

establish new corrective norms when none exist. Or to put the idea in more general terms,

when agents claim rights on the grounds of their participation in a legitimate social practice,

they undertake an obligation to the other members of that group to uphold norms that

mitigate the pathologies of that practice in order to uphold and improve its legitimacy.

We can apply this account to familial obligations, one of the paradigmatic examples

we began with. Why does the fact that two people are siblings (normally) give them special

obligations to one another? The reason is that, as members of a family, siblings presume

entitlement to certain privileges - care, attention, love, support. Along with these are privileges

of privacy. In delicate matters, members of a family can rely on the support of co-members

without publicizing their need for it, “keeping it in the family,” as the saying goes. But every

family is different, and each works out for itself how extensive these privileges are, and how

onerous the obligations of membership will be. These standards are not (typically) worked out

formally, but they are shaped through a kind of political process, as the different members

each act in the ways they interpret as appropriate for their role (whether by helping with

homework, doing chores, asking for money, throwing family birthday parties) and seeing

which of those actions get picked up, mimicked, and ultimately adopted as family practice.

This political family structure comes with its problems. Such informality invites deviance,

free-riding, and simple confusion. Who will host the next holiday? Who will house the

parents when they are suddenly unable to care for themselves? How will inheritance be split?

For many such questions, simple customs are developed to solve the problems. Holidays are

hosted on a regularly rotating basis. Ailing parents will be cared for by whichever child has

the best combination of free time, available money, and living space. Other problems demand

111

more precise solutions. How inheritance will be divided is often explicitly negotiated in the

form of a will, ultimately answerable only to the dying party. So long as a person claims a

share of the inheritance, or a place at the thanksgiving table, they have an obligation to do

their part in maintaining these rules and customs, and to working with the others to create

new rules and customs as new problems arise. By claiming these rights, family members

undertake familial obligations.

Might this account of associative obligations apply to the practice of promising as well?

Possibly.126 The idea would be that people often cannot satisfy their interests without

coordinating behavior with others, and that promising is one of the best tools we have for

arranging such coordination. There is an intrinsic tension between the need for coordinating

institutions and the rational self-interest of specific agents that might tempt non-performance.

Corrective norms would thus include strong prohibitions on promise-breaking, the right to

publicize broken promises, the privileges that come with “trustworthiness.” So when we

claim the rights of a promisee (the right to performance), or even the rights of a promisor (the

privileges of trustworthiness), we undertake the institutionally specified obligations

constitutive of a practice of promising that we, even in our own promise here and now, are

working out.127

126 Similar “practice” accounts of promising can be found in Rawls, A Theory of Justice, 344–348

and more recently Kolodny and Wallace, “Promises and Practices Revisited”. For contrary views, see Scanlon, What We Owe to Each Other, 295–327; Shiffrin, “The Divergence of Contract and Promise.”

127 This answer also illuminates the question, “Can a promise be made in the state of nature?” - a question long thought to be the final nail in the coffin of practice account of promising. If promising depends on the existence of institutions (perhaps for the sake of enforcement a la Hobbes), then promises cannot happen in the state of nature. But many theorists agree that promises can happen in the state of nature. The associative obligation view puts a new spin this affirmative answer. Yes, promises can happen in a state of nature, but only if it begets a larger practice of promising. If we make a “promise” in the state of nature, but nobody ever mimics our behavior, the promisor does not perform, and there are no costs for non-performance, then we have not actually made a promise. Instead, we are in the strange

112

The State System as a Social Practice

Might the foregoing picture explain how states are obliged to follow international law?

The suggestion would be that, like the school club, the family, or the institution of promising,

the state system is a social practice constituted in such a way that it grounds associative

obligations. Among these obligations, or perhaps at their heart, is the associative version of

Dworkin’s duty of mitigation, what I called the Associative Duty of Mitigation above. We

need only add that states would discharge this obligation by complying with international

law.

To see why this might be so, the first question is whether there is a plausible

interpretation of the state system that renders it the kind of social practice that grounds

associative obligations. Early I outlined five features that seemed, in the fictional student

club, to be jointly sufficient for such obligations. Those features might be applied to state

conduct as follows. The state system is (1) a social practice, which is (2) organized around

widely acceptable purposes and presumed to be the source of rights and privileges for

members, (3) at least partially organized by constitutive or regulative norms that fulfill a

corrective function, and, finally, (5) such norms provide appropriate grounds for criticism and

sanction among members, simply in virtue of membership.

There is a considerable case to make that the state system in fact has all of these

features. In The New Sovereignty, Chayes and Chayes argue as follows.

situation of finding out that we never actually did what we thought we were doing. We thought we were doing a thing called promising, and had promising been taken up by others, then maybe we would have actually been doing it in that first instance. However, since nobody took up the idea of promising - it never caught on - we were not actually promising in that first instance. It is as if we were trying to start a new dance craze, “The Pilch”, by gyrating frantically on the dance floor but nobody followed suit. Ever. Were we doing The Pilch? Not really. We were just gyrating frantically.

113

That the contemporary international system is interdependent and increasingly so is not news. Our argument goes further. It is that, for all but a few self-isolated nations, sovereignty no longer consists in the freedom of states to act independently, in their perceived self-interest, but in membership in reasonably good standing in the regimes that make up the substance of international life…. Sovereignty, in the end, is status — the vindication of the state’s existence as a member of the international system. In today’s setting, the only way most states can realize or express their sovereignty is through participation in the various regimes that regulate and order the international system.128

Alexander Wendt has famously expanded on similar ideas, as exemplified by the following

passage:

Sovereignty is an institution, and so it exists only in virtue of certain intersubjective understandings and expectations; there is no sovereignty without an other. These understandings and expectations not only constitute a particular kind of state — the "sovereign" state — but also constitute a particular form of community, since identities are relational. The essence of this community is a mutual recognition of one another's right to exercise exclusive political authority within territorial limits. These reciprocal "permissions" constitute a spatially rather than functionally differentiated world — a world in which fields of practice constitute and are organized around "domestic" and "international" spaces rather than around the performance of particular activities.129

Wendt’s writing style can be unfamiliar for analytic philosophers, but his idea is the very one

captured by Chayes and Chayes: States are members of a group, the society of states. They

owe their rights — sovereignty, the right to rule — to that group such that without the

recognition of the other members, the state would not be the bearer of those institutionally

specified rights.

Practice affords us plenty of examples of states that are sovereign only because they

are recognized as such by an international society of states. Israel, certainly in the earliest

parts of the post-war period, is a perfect example where domestic procedures for foundation

and the popular creation of a sovereign state mattered far less, if at all, than international 128 Chayes and Chayes, The New Sovereignty, 27. 129 Wendt, “Anarchy Is What States Make of It,” 412; see also Kratochwil, Mansfield, and Ruggie, “International Organization.”

114

insistence that such state be created and respected. South Sudan and South Korea (and

probably North Korea, though this case is less clear in light of its roguish behavior) are also

probably examples of countries that enjoy sovereignty not because of some domestic process

whereby citizens gave up their rights, and not because of a dictum from God establishing their

rights, but because the international community recognized protective associations in those

regions as sovereigns.

This “exogenous” view of sovereignty also explains cases of state-like agents that are

not actually or clearly states. A simple example is the Principality of Sealand, a manmade

island of the eastern coast of England that claims sovereignty and political autonomy.

Although citizens of this principality have reached complete unanimity about their

independent status, no state in the world recognizes their sovereign rights. As an interpretive

matter, it would be an error to claim that Sealand is a state. A more recent and serious

example is the Autonomous Republic of Crimea, who supposedly seceded from Ukraine in

February 2014 after a landslide vote to do so.130 At the moment, most European countries and

the Unites States reject this vote as illegal, thus refusing to recognize Crimea’s sovereignty.

Russia, on the other hand, has already treated with Crimea, thus implicitly recognizing it.

Thus the exogenous view would tell us that Crimean sovereignty is unsettled, as is the case in

practice.

These authors also describe the state system as a political practice — one whose terms

are worked out as a matter of proposals made by peers and then instituted through a process

of uptake, repetition, and mimicking. Chayes and Chayes point out that states frequently

discover the boundaries of their treaty obligations by “testing” the limits of what cosigners will

130 Of course, the legitimacy of this vote can be questioned both on the basis of how the options on the ballot were phased and also on the basis of large social group who opposed secession boycotting the elections.

115

tolerate.131 In a more cynical state of mind, we might take this as an example of states trying to

cheat on their agreements, but nothing about the case forces this interpretation on us. In fact,

given that states willingly enter into agreements from a shared sense of urgency, cheating

seems not to be the best interpretation of such envelope pushing. Instead, Chayes and Chayes

suggest, this testing is a way of specifying the terms of membership — in this case,

membership in a treaty — by seeing which actions will be tolerated and which will not. State

actions, therefore, are a kind of legislative proposal about the boundaries of permissible state

action, to be approved or rejected by the others as indicated through their reactions.

An example will be instructive. Consider the international legal custom of flying high

orbit spy planes over other countries.132 Prior to the cold war, international legal custom had

been that state boundaries rose straight from the Earth’s surface to the end of the atmosphere.

However, when the US and USSR began flying spy planes over one another, other members

of the international community generally accepted this. Gradually, as other countries

developed the technology to do so, they followed suit. This was very clearly a case of

presumed legal obligations changing through a political process of performative proposal, of

pushing the envelope, and the subsequent development of new customs through mimicry and

further boundary testing.

These are especially clear cases because they are extreme, but the image of the state

system as a social practice lived by highly autonomous states but governed by norms and

organized around widely acceptable purposes is apparent even in humdrum cases of

international political life. As Dworkin says, and as custom assumes, states are indeed

sovereign. They are politically independent units, afforded vast rights of non-interference and

autonomy. But those rights have readily apparent limits. States are not supposed to engage in 131 Chayes and Chayes, The New Sovereignty, 12. 132 Cheng, “United Nations Resolutions on Outer Space.”

116

aggressive warfare, and, for the most part, they do not. In the event of natural catastrophes,

states are expected to send aid and support, and, for the most part, they do. States respect

borders, elections, trade deals, norms about he treatment of diplomats, standards about how

violent conflict will be conducted, and when they do not they are open to public shame and

formal sanction by the others.

And the purpose of such a system? That is an interpretive question, and different

people might give different answers, but a few suggestions do not seem too controversial. We

forbid aggressive war, put very high standards for the permissible use of interstate violence,

and give special privileges to diplomats because we want peace, or at least security. Even if we

ourselves do not care so much about this purpose, we can understand why someone

reasonably might. Indeed, as Dworkin says, one of the major reasons for the creation of the

balkanized Westphalian state system in the first place was to substitute “ economic

competition for the bloody religious conflicts that had marked the previous century.”133 We

allow states to extract their domestic resources and to refine and sell them abroad for the sake

of amassing wealth because we care about prosperity. We expect states to respect borders and

territorial integrity, to respect domestic legislative procedures, and not to dominate in joint

ventures because we care about the preservation and toleration of diversity. I may be wrong

about what the purposes of the state system are — maybe we care about autonomy rather

than diversity, for example — but surely some such account is correct. What is important for

present purposes is that something like these purposes — values that are widely acceptable

from the internal perspective of the members — are a necessary part of making sense of the

state system that we have.

133 Dworkin, “A New Philosophy for International Law,” 16.

117

As Wendt emphasizes, these norms of the state system play a double role. On one

hand, they regulate the behavior of states in fact. That is why, for example, states act

differently with respect to events within their borders as opposed to events without. The US

sent aid to the Philippines after Haiyan, but sent the National Guard and FEMA to Louisiana

after Katrina. Our government creates rules about how our elections will go, but does not in

Lithuania. Why? Because there are rules about what states are allowed to do domestically, and

what they are allowed to do internationally, and these rules are different.

Aside from their regulatory function, these norms also specify what it is to be a state. A

state is the kind of thing that has the right to hold and regulate domestic elections, and to

exclude foreign governing bodies from interfering in those processes. States have the right to

protect citizens from natural catastrophes, and a responsibility to do so that ranges far beyond

similar responsibilities to non-citizens. The rules that govern what states may and may not do

also specify the rights that define what it is to be a state in the first place. In this way, the state

system is not only presumed to regulate states, it is the presumptive source of those rights that

set an association of persons apart as a state.

As was discussed in Ch. 3, the state system practice is prone to pathologies. Its

decentralized structure makes problems of common concern, the solutions to which demand

coordination, more difficult to solve than in a hierarchical system where subjects could simply

defer to the will of a Hobbesian sovereign. The problems of the Kyoto Protocol and the

recurring challenges to the legitimacy of international courts are examples. Norms of non-

interference put minorities at risk of domestic oppression. Norms that make state autonomy

and self-help the default presumption inhibit transparency and communication, undermining

assurances of security and non-aggression.

118

So the final question, then, is: Is the state system at least partially constituted by norms

that, if complied with, could correct for these pathologies? Again, I submit that the answer is

yes. Those norms that are both constitutive and corrective are international law. Through a

decentralized political practice, states have worked out a set of regulatory norms that are

specially capable of addressing the negative externalities of that very practice. Many of these

norms take the form of customs. Diplomats are presumed to have special rights and

privileges, because they help states communicate their intentions and to give assurances for

effectively. Documents that specify the boundaries of permissible treatment of individual

humans (UDHR) or laborers (ILO standards) are treated as if they have a special reason-giving

status for states, even though not all states got to take part in their enumeration. A modern

version of the idea of the treaty emerged, whereby states could formally negotiate mutually

acceptable terms for coordination on specific issues and make those terms explicit and

publicly available.

What we have in the state system, therefore, is a social practice, guided by widely

acceptable purposes, presumed by practitioners to be the source of rights and privileges,

prone to pathology, and at least partially constituted by norms that, with general compliance,

fill a corrective function. These norms also seem to be an appropriate basis for sanctioning

non-compliant member states. In terms of the five criteria discussed earlier, this system is

isomorphic with the school club, the family, and perhaps even the institution of promising.

So, just as members of the club claim the privileges of membership or members of the family

claim familial rights, when states claim the rights of sovereignty — rights institutionally

specified by the political practice that is the state system — they undertake an associative

obligation to do their part in mitigating the pathologies of the system that specifies and grants

those rights.

119

The Obligation to Mitigate and Compliance with International Law

It is tempting to stop here and conclude that, because states have an obligation to

mitigate the pathologies of the state system, and because they can do this by conforming to

international law, they therefore are obligated to comply with international law. But this

argument is open to an early and simple objection. Even if we agree that states have an

associative obligation to mitigate, and we agree that compliance with international law is one

way that a state can discharge this obligation, we have yet to see why states have an obligation

to comply with international law. There may be other ways, aside from acting in conformity

with international law, for states to address pathologies and thereby discharge their

obligation. It follows, then, that states do not have an obligation to comply with international

law per se, so long as they realize one of the alternatives.

What might such an alternative be? The answer is not obvious. In theory, one route

would be to work towards the dissolution of the state system itself. If states’ obligations are to

get rid of the problems that come with the state system, one way to do this is to get rid of the

state system altogether in favor of international anarchy. Even if this needn’t amount to a

return to the state of nature for individual humans — they can remain within their respective

protective associations (formerly states) — it would be a return to the state of nature among

protective associations. We need not assume that this would make human life solitary, poor,

nasty, brutish, or short, though it certainly would mark a dramatic change.

Even if this is conceptually coherent, it is not obvious that any such dissolution would

in fact rid us of the state system’s many pathologies, rather than simply obscure them. Wendt,

for example, argues that what might appear to be the dissolution of the state system into

anarchy is really nothing more than the assertion of stronger rights to autonomy and self-

120

governance within a state system.134 If the leaders of the world come together and say, “No

longer shall we coordinate our behaviors through this political practice. Instead, there are no

rules about how protective agencies shall treat one another,” this would itself be an example

of political legislation within the state system, if not in name, then in real practice. As agents

subsequently competed for resources or whatever agents do in the state of nature, they would

be living out the rights that they give one another as members of a political community. Just

trying to think what it would mean to dissolve the state system is conceptually problematic, at

least short of a return to all-out anarchy, in which the life of man really might be, as Hobbes

suggested, solitary, poor, nasty, brutish, and short.

We can take this one step further. Let us assume that we can make conceptual sense of

the dissolution of the state system into anarchy. In the first place, there are forceful moral

arguments why this ought not to be done. Importantly among these is an argument on the

basis of the values that constitute the purposes of the state system that we would be foregoing

by dissolving it. I have suggested that the state system is organized towards the achievement

of security, prosperity, and toleration. What, by comparison, would be the worthy purpose

served in returning to the state of nature? To avoid legal encumbrances? To avoid

burdensome duties to aid others? More charitably, one might say that the purpose of a return

to anarchy is the pursuit of freedom. But what is liberty without security, even for a state?

What is freedom without prosperity? For there is no reason to assume that these go hand in

hand. The valuable purposes of the state system cannot be achieved in a state of nature, and

the values that might motivate dissolution do not outweigh them.

The final, and perhaps most obvious response to the anarchist, is that there states have

no practicable avenues to bringing about the dissolution of the state system. Even if

134 Wendt, “Anarchy Is What States Make of It,” 391.

121

dissolution was conceptually coherent, and even if it was not open to decisive moral objection,

it still would not be practically possible. Partly this is due to political will, or the lack thereof.

People and states genuinely benefit by being part of the state system. Especially in the post-

war period, we, as a world, have made significant gains precisely in the areas of security,

prosperity, and toleration. In many respects, the promises of the modern state system are

being gradually fulfilled.

A different alternative would be for states to contribute to the establishment of a

centralized world state. Most, if not all, of the pathologies that Dworkin, Beitz, Kumm, and

James identify arise from the decentralized character of the modern state system.

Decentralization leads to coordination problems. Excessive autonomy leads to risks of

domestic oppression and unequal distributions of wealth. If the problem is not governance per

se as the anarchist thought, but its decentralization, then perhaps a state could satisfy its

obligation to mitigate by helping to bring about a centralized world state that would, almost

analytically, avoid these pathologies.

This alternative seems not to suffer the conceptual shortcomings of the anarchist.

While administrating a world state would surely be a herculean task, there is not, on the face

of things, any conceptual incoherence in the idea of one big state. Wendt might disagree and

claim, as we have seen, that there is no state without an other, but this appears to be a point

about semantics.135 Call it what you will, state or global protective association, it seems at least

conceivable that such an organization could come to be.

Working towards a global state also at least conceptually could avoid the moral

problems faced by anarchy. Surely a global state could pursue the same values that constitute

the purpose of the modern state system. We might argue about whether a global state will be

135 Wendt, “Anarchy Is What States Make of It”; Wendt, Social Theory of International Politics.

122

the most efficient way to achieve certain values — toleration of diversity comes to mind as the

kind of thing that might be quashed by bureaucracy — but nothing in the practice account of

associative obligations says that states have an obligation to correct pathologies in the most

efficient way, even if this proves to be the case. But here, as is often suggested, practicability

is a serious problem. Even if the political will were present internationally to create a world

state (which it is not), we lack the administrative capacity. We lack a shared history of

acceptable world governance or even of acceptable governance in the first instance. What we

do share is a long tradition of self-governance and autonomy, and a world state marks a

dramatic break from that. Perhaps we can see a world state on the distant horizon, but it is, I

believe, excessively optimistic to think that there is anything states can do now to hasten its

arrival.

A third alternative, intermediate between the two extremes of international anarchy

and a world state, would be a form of state exceptionalism.136 Rather than depending on

cooperation among juridical equals, a special set of exceptional rights and duties could be

granted to especially powerful countries. With these special liberties, an exceptional state

could do the work of mitigation unilaterally, while non-exceptional states are expected to

simply comply with law. If such arrangements would best mitigate problematic tendencies of

the larger state system, then it would seem to follow that law is not authoritative, or at least

not in a fully general way. It would not simply follow that the exceptional states can often

justifiably set aside or re-interpret otherwise content-independent reasons to comply with

standing law. They would lack such reasons in the first instance.

This proposal is conceptually coherent and practically feasible. It is coherent in the

sense that my account allows for the suggested possibility, at least in theory (perhaps much as

136 A recent version of this familiar position is advanced by Ku and Yoo, Taming Globalization.

123

in the Wild West, when the town sheriff is seen as above the law, if only to provide protection

against outlaws). It is also practically feasible, in the sense that it is a practice that assigns

special privileges stands a chance of being more effective than a practice in which all states

normally follow the law unless engaging in performative legislation. Because most states are

still expected to comply with the law, perhaps many or most of the most important

coordination problems would still be solved—at least so long as public non-compliance by the

powerful few did not destabilize general cooperation.

There is nevertheless reason to be very skeptical about this proposal under

contemporary conditions, for moral and practical reasons. For one thing, such special

privileges may be open to substantive moral objection. For example, the concentration of

legal power in the hands of a single state or a small group of states may invite the

marginalization of ways of life and private conceptions of the good among less powerful

societies. Power concentration invites social domination, which is inconsistent with the state

system’s basic purpose of preserving diversity (though perhaps mainly for the sake of

security). And if non-exceptional states do feel dominated, one can doubt whether they will

comply with international law for very long, instead of invoking exceptions for themselves.

To the extent such a system does not enjoy stable even if rough compliance, it is less likely to

successfully fulfill its required corrective role.

This suggests a further, largely practical line of objection. It is highly doubtful, in real

practice, that any single current state is in fact sufficiently powerful or wealthy to unilaterally

resolve the pathologies of the state system, in which case no state can claim exception to the

laws that generally apply. Indeed, as Chayes and Chayes explain, “effective sovereignty”

under contemporary conditions, even for the most powerful states, is less a matter of taking

124

exceptions than of “capabilities” to facilitate cooperation, less a matter of “going it alone” than

of enlisting other states in the service of common goals.137

This is still not quite to say that international law is authoritative for all states in the

same way. But we can see why this is a defensible thesis by considering the difficulties of the

alternatives. One possibility is that exceptions to general law be assigned to different states on

an issue-by-issue basis, along with obligations to lead unilaterally and set terms of

cooperation for other states. But it plainly won’t work, say, for China to be in charge of

coordinating a response to climate change, for the US to be in charge of coordinating nuclear

disarmament, and for Russia to be in charge of reducing the risk of bad leaders from coming

to power. Not only are such pathologies all wound up with one another, such a scheme would

create unworkable coordination problems, which would only erode the system’s legitimacy.

Nor, it would seem, could mitigation labor be divided, not by issue, but regionally (e.g., with

China in charge of mitigation in one part of the world, Russia in another, the US in another.)

There is ample room for regional leadership without violating international law, and it is not

clear how this would successfully manage pathologies that arise within the state system across

different regions.

More generally, it does not seem that exceptional rights could be granted to any other

subset of states without raising serious questions about the workability of the system, at least

given the dynamics of state recognition. How would states be given these exceptional rights?

Who should have them, and on what basis? If Wendt is correct, the answer to both questions

is not practical but conceptual: any rights are given on the basis of recognition by other states,

within the community of states that already recognize one another, on the basis of the reasons

recognized by the community. But if no state is powerful enough to somehow command

137 Chayes and Chayes, The New Sovereignty, 197–201.

125

recognition as the only exceptional state (perhaps as the sole guarantor of security), which

states would be permitted exceptional status would be a matter of mutual recognition. A

natural arrangement is a roughly uniform system, in which no state is afforded general

exceptions, except as specified by specific rules, or by a proposed re-interpretation, and it is at

best unclear why any less uniform system would generally do a better job at mitigation.

One final alternative should be addressed. I say that states ought to comply with

international law, and that this obligation is importantly tied to mitigation. But it is unlikely

that every individual legal rule fills this function. What would be wrong, the challenge goes,

with an international legal regime that allowed states to pick and choose the legal rules that

they will follow on the basis of good-faith judgments about whether those rules help mitigate

evident pathologies?

Such a legal regime might be justified, but again, I believe that there are pragmatic

reasons to doubt this. A legal system that allows states to pick and choose the laws they want

to follow, as an empirical matter, will not be able to sustainably fulfill the corrective function

envisioned. Not only would such a system be prone to abuse, it puts states back in a

coordination problem akin to the ones that law was meant to solve. Worse, because current

legal rules are interconnected – that is, they genuinely for a legal system – ignoring some rules

may undermine the ability of other parts of the law to correct.

The practice approach can avoid this with the following addition: So long as

compliance with some subset of currently recognized laws would correct pathologies, then

states have the same obligation to comply with all recognized laws. This way, states have a

duty to comply with international law qua law, not because each individual rule corrects for

pathologies, but because only when the entire system is treated as authoritative can it, as a

whole, do so. However, as has been discussed, because international law is open to political

126

forms of legislation, states can amend or even discard laws that do not fulfill this function over

time. This point is important to the explanation of the content-independence of (all or most

of) international law as we find it.

Perhaps there are further ways that a state could contribute to the mitigation of the

pathologies of the state system besides compliance with international law, dissolution into

anarchy, the establishment of a world state, or the establishment of an exceptionalist system.

Short of answering each possibility, let me simply suggest the following. If a state does not

wish to comply with international law while at the same time claiming the rights of

sovereignty, any excuse for noncompliance must be of the form: In order to satisfy my

obligation to mitigate, I will do X, where X is inconsistent with compliance with international

law, but also (1) contributes to the mitigation of the pathologies of the state system, (2) pursues

purposes of comparable or greater value than the purposes of that system, and (3) is both

conceptually and practically possible. In that case, in the apparent absence of alternative

means for mitigating the pathologies of the state system, compliance with international law is,

at least presumptively, the only means available to states to satisfy their obligation.

This, then, is the answer to Dworkin’s doctrinal question. Why should states follow

international law? When states claim the institutionally specified rights of sovereignty, they

undertake obligations to mitigate the pathologies of that system. In the apparent absence of

practicable and morally unobjectionable alternatives, this obligation can only be fulfilled by

complying with existing corrective norms, amending existing norms so that they better fill the

corrective function, and legislating new norms through the political means available to

members when no corrective norms exist. At the moment, those existing corrective norms are

the rules that constitute the international legal system, and the means for amending and

creating new norms are what states currently recognize as the sources of international law —

127

custom, treaties, judicial rulings, and administrative commands. Thus states have an

obligation to comply with international law.

Associative Obligations and Authority

I have claimed that several different views — natural law, state consent, and Dworkin’s

New Philosophy — fail to capture the sense of authority presumed in contemporary

international practice, which I have characterized as three characteristics. First, international

law gives states particularly a specific kind of reason for action that is distinguishable from

reasons given by advice and requests. Second, these reasons are normally decisive in practical

deliberation, but can sometimes be overridden or defeated by countervailing considerations.

Third, international law gives states such reasons for action even when specific legal rules are

inconsistent with widely accepted moral truths. Together, these three conditions represent

the form of international legal authority as presumed in practice. Reflections on state consent

theory raised a fourth criterion for a successful account of international law as we find it: that

it explains how a maximally large set of the legal rules currently recognized in practice could

have the kind of authority those rules are presumed to have. This is, I suggest, the scope of

authoritative international law.

Does the practice approach capture both the form and scope of international law as we

find it? In part, this is an empirical question that turns on what tends to fulfill the corrective

role. But there is reason to suspect that the legal rules authorized by the practice approach do

correct better than any of the considered alternatives. This is particularly clear in the case of

state consent. Because it would demand such vast revisions of existing international law, its

narrowness likely excludes many important corrective norms such as the ILO’s labor

standards and possibly even human rights law. While natural law theories can avoid

128

problems of scope (by accepting the challenge of giving a moral justification for apparently

immoral laws), they insist on thinking of legal duties as a kind of moral duties. This shuts

down the ambition to find an account of law that speaks to diverse agents on grounds that

each can accept from its own perspective. By grounding legal duties in associative obligations,

and by linking sovereign rights to compliance, the practice approach constructs a form of

political and legal address that can use a state’s own reasons to justify the demand for

conformity. Again, which of these forms of legal authority best fulfills the corrective function,

assuming both are available, is an empirical question. But these alternatives represent the

state of the art in contemporary international legal theory, so even this merely relative success

for the practice approach is significant.

How does the practice approach explain the reason-giving characteristic of

international law? There are two ways to answer this question, one that answers from an

external perspective, and one that answers from an internal perspective. It is not obvious that

both are necessary — the external suffices — but the internal can be instructive as well. The

external explanation is just to say that states have an associative obligation to comply with law

as a result of their claiming the institutionally specified rights of sovereignty. We have now

discussed this at length. The internal explanation makes reference to a state’s own reasons for

claiming those rights. The strength of the state’s claim to those rights is limited to the right of

the institution to shape and give those rights, so if the state has reason to claim the rights, the

state has reason to be concerned about the legitimacy of the system. From there we can make

the now familiar arguments about the connection between legitimacy and pathologies and the

corrective role of international law to explain, on the basis of the state’s own reasons, why it

ought to comply with international law.

129

But we have seen that it is not enough to show that law is reason giving, but that these

reasons are somehow distinct from the kinds of reasons given by advice and requests. How

does the practice approach make this differentiation? Again, there are two answers for this.

First, associative obligations are the appropriate grounds for accountability seeking behaviors

upon non-compliance. When one shirks one’s associative obligations — as in the case of the

brother who skips his sister’s wedding or the promisor who reneges — one is rightly

sanctioned or rebuked proportionately to the gravity of the offense. Failure to grant requests

and especially to heed advice (at least when it is good advice) may result in bad consequences,

but these would not be rightly interpreted as sanction on rebuke. In the normal case, it is not

wrong to ignore advice or to refuse requests. It is wrong, however, to fail to fulfill associative

obligations, which on the practice account explains why it is wrong to violate the law.

Second, when somebody requests that I do some action, or advises me to do some

action, there is no sense in which I owe him or her my performance. According to the practice

approach, conformity with associative obligation is owed to others. In the case of the familial

obligation, I owe my conforming performance to the members of my family. In promising, the

promisor owes performance to the promisee. In international law, the state owes compliance

to the other states whose respective claims to sovereignty also depend on the legitimacy of the

system and the mitigation of pathologies. To use Margaret Gilbert’s term, legal obligations are

directed, whereas reasons to heed advice or grant a request are not.138

Next, how does the practice approach explain the defeasibility of law? By grounding

legal obligations in associative rather than moral obligations, the practice approach shows

how law and morality can conflict. The analogy with promises is instructive. We are, I believe,

bound by our immoral promises. If I promise you that I will drive the getaway car to help you

138 Gilbert, “Is an Agreement an Exchange of Promises?”.

130

rob the bank, but then drive off just as you step through the front doors (perhaps because my

conscience gets the better of me), then it seems that you are in a position to hold me

accountable. After all, I just broke a very important promise to you! That does not mean that I

should have kept the promise. All things considered, I did the right thing to drive off. Not only

did I keep from doing something immoral myself, I made it much easier for the authorities to

apprehend a known criminal. All of this is just to say that associative obligations, the kind of

obligations at the heart of the practice approach to international law, can be defeated by

sufficiently grave countervailing considerations. Law is no exception.

How does the practice approach explain the content-independence of legal authority?

Given the current absence of practicable or otherwise better alternatives, the only or at least

best way for states to discharge their obligation to mitigate the pathologies of the state system

is to comply with international law, whatever it happens to be. This means that even if

particular legal rules conflict with widely accepted moral truths, states still have an obligation

to conform.

Consider again NAFTA, an example of morally problematic but presumptively

authoritative international law. This trade agreement is having the double effect of

exacerbating wealth inequality between Mexico and the Unites States while simultaneously

displacing traditional Mexican farmers who now cannot compete with commodities from the

US. The obvious objection to the authority of this law is that it is immoral, but the practice

approach maintains that states ought to obey it despite this immorality because international

law fills a corrective function. But why should this be so?

The answer shifts attention from NAFTA itself to the customary legal principle pacta

sunt servanda. Many of the problematic tendencies of a politically decentralized practice such

as the state system have to do with failures of coordination. Like promises and agreements for

131

individual persons, treaties are one of the most powerful, effective, and widely recognized

solutions to coordination problems. But they only work when both parties have reasonable

assurance that their agreement will be honored. The practice approach make sense of why

international law would include a strong customary prohibition on the violation of treaties,

even when they are immoral. Treaty making only solves coordination problems only when

there is assurance of performance, and the strong prohibition on reneging, even if the

agreement seems to either or both parties to be inconsistent with moral truths, backed up with

the force of law gives that assurance. States ought to comply with NAFTA, despite its apparent

immorality, because there is a justified legal principle, of presumed compliance, that

commands conformity to treaties, and that legal principle is justified by its corrective role.

Perhaps this can justify immoral treaties, but can an analogous argument be offered

for immoral customary or administrative law? Yes, an analogous argument can be given. As

with treaties, these source of law need to be able to give states assurance that others will

comply in order to fulfill the corrective function by coordinating state action. Such assurance

can only be provided by a strong prohibition on non-compliance, even when the terms of the

law are deemed morally problematic by one of the parties. In fact, the above example of

NAFTA actually makes this very point, since pacta sunt servanda is best understood as a

custom that sometimes has morally problematic consequences but is nevertheless backed up

by the force of law. States need, in a sense, to coordinate on a means for facilitating future

coordination. That states have, in fact, settled on the customary principle pacta sunt servanda

fulfills this coordinative need.

But the practice approach goes further: states have a duty to comply with law qua law

because only then can the legal system as a whole fulfill its corrective function. If states were

legally allowed to pick and choose which rules they would follow, the legal system could not

132

serve to reliably coordinate state behavior in the current decentralized state system and

corrective legal rules could lose the institutional support of other non-corrective rules that

they depend on. Thus international legal authority, on the practice approach, is truly content-

independent – independent even of whether the content of an individual law itself corrects for

some pathology. Of course, this forces the following qualification: absent other special

justification, any particular international legal rule is only as authoritative as the going version

of the international legal system is capable of mitigating pathologies. This sheds further light

on defeasibility: the better the legal system as a whole corrects for pathologies, the more

reason states have to follow it and the fewer countervailing concerns suffice to defeat that

obligation.

To balance this controversial point about content-independence, the practice

approach adds that the state’s obligation to mitigate also includes an obligation to amend

existing law to better fulfill the corrective function and an obligation to create new law when

none already exists. Thus the practice approach leaves open the possibility that consistency

with widely accepted moral truths might be part of how a revised version of an existing legal

rule “better” fulfills its corrective function. So this approach need not be completely divorced

from morality, even though legal obligation is not ultimately a moral obligation as might be

found in the Kantian or Lockean traditions.

This raises the question: How revisionistic must the practice approach be? The answer,

I suggest, is “not at all.” The practice approach holds that states have an obligation to comply

with international law as they find it. This duty varies in how defeasible it is, depending on

how well the going legal system, as a whole, corrects for the pathologies of the state system. Of

course, as just mentioned, states may deviate from existing law — that is, they may act

contrary to currently recognized norms — if they believe that the new norm instantiated by

133

their action could be accepted by others as an amended and improved version of existing

norms. But in a political practice such as the state system, proposing new norms is done

performatively, by acting in the way that you think members out to act. Whether or not this

performative legislative proposal is accepted or not — and so whether the action was legal or

not — depends on whether other members follow suit. So deviation is risky. And when it does

not manifest in sanctions for non-conformity, it is because the supposedly deviant behavior

became the new norm. But because the practice account does not insist on consistency with

widely accepted moral norms, we are not forced to revise immoral laws. Because it does not

insist on the willed acceptance of legal rules, we do not have to exclude legal rules that states

do not consent to. And because the practice approach does not favor one source of

international law — treaties or customs or judicial rulings — it can accommodate all currently

recognized sources. If the practice approach is revisionistic, it is only to the extent that states

can identify, on publicly acceptable bases, new rules that would full currently unaddressed

corrective needs and new versions of existing rules that would better correct for pathologies.

But these revisions are far less sweeping than those proposed by any of the alternative views

considered.

Concluding Remarks

What the practice approach offers, in the end, is a theory that uses a familiar and

everyday kind normativity — associative obligations — to ground an account of why states

ought to comply with international law. It does not insist that practitioners change their

concept of legal authority, but leaves the current conception intact. It does not insist on vast

revisions of existing law, but illuminates both the conditions under which revision would be

134

appropriate, and how such revision might be achieved. It offers an analytically rigorous

account of why states ought to comply, while also giving guidance about how to link demands

for compliance with reasons that actual states otherwise have. It integrates Dworkin’s highly

promising New Philosophy, including much of what makes it appeal to theorists who situate

themselves within the natural law tradition, with the positivist commitment to the separation

between law and morals, and offer an answer to the doctrinal question of why states should

obey international law. For these reasons, I believe that the practice approach represents a

significant step towards understanding international law and, perhaps, law in general.

135

CHAPTER 5: Conclusion

It was late in the process of writing this dissertation that I was asked what has become,

to me, the most challenging question: Is the practice approach merely a post hoc justification

for a bitterly unjust world order?139 In my view it is not, and not because I disagree that the

world is unjust. But given my conservative methodology and the fact that my criticisms of

competing views frequently stem from what I apparently deem excessive revisionism, when a

friend first asked me this question, I realized that I would need to answer it. Remarkably,

reflection on a question that I initially saw as a challenge has shown me yet more reason to be

attracted to this view.

Here is one way to think about this project: Traditionally, scholars have asked, is

current legal practice justified? To answer, these scholars summon their preferred moral or

political view, and then see how current practice matches up to that. My project asks, is

current legal practice justifiable at all? Whatever my preferred moral and political views are, I

know that there are many more out there. What I want to know is, does any one of them serve

to justify the body of rules that constitute international as we find it? In fact, I wanted to go

beyond just considering presently acknowledged moral and political views, but to ask

whether such a view could even be imagined, even if it is a view that nobody, in fact, holds.

There are at least two ways to interpret such an investigation. The first, and the one

that motivated my friend’s question, comes from a sense of dissatisfaction with the status quo

and perhaps the suspicion that I – a straight, white, male, American who has profited

enormously from the unjust distribution of privilege in our world – am trying to rationalize

that privilege, consciously or not. My work, on such an interpretation, is a kind of perverse

139 My thanks to Megan Zane and Amanda Trefethen.

136

and desperate contortion of morality in the service of subtly maintaining even the pathologies

that I identify. Given the opacity of so much of our psychology, I hope that this is not what I am

up to.

The second interpretation, thankfully, is very different from the first. It too comes from

a sense of dissatisfaction with the current state of international politics, but adds to this a kind

of humility about my private conception of the good and deference to other equally

reasonable conceptions. I am not so secure in my moral convictions that I am prepared to

prescribe the exertion of the force of law when they demand it, especially if reasonable people

around me do not share my certainty. But there is no stronger condemnation of a legal rule

than to say that no imaginable moral or political view could possibly justify it, and that is

precisely the form of critique the conservative methodology of this view can ground. If a law

cannot be interpreted consistently with a view whose explicit purpose is to maximize fidelity

to going practice while remaining logically coherent and meaningfully connected to widely

acceptable values, then practitioners will have identified a locus for agreement, for shared

purpose, and for advancing the grand endeavor of international law.

In developing the practice approach, my hope has been to offer a structure for critical

reflection about international law – one that connects laws with their measurable impacts on

human life, and that grounds a particularly potent form of objection. Most importantly, it can

serve as a roadmap for collaboration across disciplines. Too often, our experience of other

fields - maybe after a talk or auditing a class – can be summed up in the following sentence:

“It was interesting, but I’m not sure how I can use it for my work.” The practice approach is

my suggestion how. On one hand, the practice approach depends on theorists who can make

interpretive judgments about the purposes of patterned social interaction, and value

judgments about which of those purposes are acceptable. On the other, it depends on

137

empirical investigation into the impact of legal rules, surveys about how people think about

law, and technological knowledge about what kinds of alternatives are available. Such

ventures outside of philosophy departments will, I think, be necessary in the near future for

the perpetuation of our discipline, but promise to nothing except to enrich our work if we can

learn to be humble enough to wonder again.

138

BIBLIOGRAPHY Anscombe, G. E. M. “On the Source of the Authority of the State.” In Authority, edited by

Joseph Raz. Oxford: Oxford University Press, 1990. Anscombe, Gertrude Elizabeth Margaret. “Rules, Rrights, and Promises.” Midwest Studies in

Philosophy 3, no. 1 (1978): 318–23. Aquinas, Saint Thomas. Summa Theologica, Part I-II, n.d. Aristotle, and Terence Irwin. Nicomachean Ethics. 2nd edition. Indianapolis, Ind.: Hackett

Publishing Co., 1999. Arneson, Richard J. “The Principle of Fairness and Free-Rider Problems.” Ethics, 1982, 616–33. Asad, Talal. On Suicide Bombing. 1St Edition edition. New York: Columbia University Press,

2007. Beitz, Charles R. The Idea of Human Rights. Oxford University Press, USA, 2011. Blake, Michael. “Distributive Justice, State Coercion, and Autonomy.” Philosophy and Public

Affairs 30, no. 3 (2001): 257–96. Bodansky, Daniel. “Legitimacy in International Law and International Relations.” In

Interdisciplinary Perspectives on International Law and International Relations: The State of the Art, edited by Jeffrey L. Dunhoff and Mark A. Pollack. New York, NY: Cambridge University Press, 2013.

Brierly, James Leslie. The Basis of Obligation in International Law. Oxford: Oxford University Press, 1958.

Buchanan, Allen, Samantha Besson, and John Tasiolas. “The Legitimacy of International Law.” In The Philosophy of International Law. Oxford: Oxford University Press, 2010.

Cassese, The Late Antonio. International Law. 2nd ed. Oxford University Press, USA, 2005. Chayes, Abram, and Antonia Handler Chayes. The New Sovereignty: Compliance with

International Regulatory Agreements. Harvard University Press, 1998. Cheng, Bin. “United Nations Resolutions on Outer Space:‘Instant’ International Customary

Law?” INDIAN J. INT’L L. 5 (1965): 23. Cohen, Joshua, and Charles F. Sabel. “Global Democracy.” NYUJ Int’l. L. & Pol. 37 (2004):

763. D’Amato, Anthony. Concept of Custom in International Law. Ithaca, N.Y: Cornell University

Press, 1971. Denza, Eileen. “Vienna Convention on Diplomatic Relations.” United Nations Audiovisual

Library of International Law, 2009, 1–7. Dworkin, Ronald. “A New Philosophy for International Law.” Philosophy & Public Affairs 41,

no. 1 (2013): 2–30. ———. Law’s Empire. Belknap Press of Harvard University Press, 1986. ———. Taking Rights Seriously. Harvard University Press, 1978. Finnis, John. Natural Law and Natural Rights. Oxford University Press, 2011. ———. “On Reason and Authority in Law’s Empire.” Law and Philosophy 6, no. 3 (December

1, 1987): 357–80. Finnis, John Mitchell. “Authority.” In Authority, edited by Joseph Raz. Oxford: Oxford

University Press, 1990. Fuller, Lon L. “Eight Ways to Fail to Make Law.” In The Morality of Law. Virginia: Yale

University Press, 1964. Gilbert, Margaret. A Theory of Political Obligation: Membership, Commitment, and the Bonds of

Society. Oxford University Press, USA, 2006.

139

———. “De-Moralizing Political Obligation.” In Joint Commitment. New York, NY: Oxford University Press, 2014.

———. “Is an Agreement an Exchange of Promises?” The Journal of Philosophy 90, no. 12 (December 1, 1993): 627–49. doi:10.2307/2940815.

———. On Social Facts. Princeton University Press, 1989. ———. “Three Dogmas About Promising.” In Joint Commitment. New York, NY: Oxford

University Press, 2014. Goldsmith, Jack L., and Eric A. Posner. The Limits of International Law. Oxford University

Press, USA, 2006. Griffin, James. On Human Rights. Oxford University Press, USA, 2009. Gritten, David, Lucy Rodgers, and Emily Macguire. “Syria: The Story of the Conflict.” BBC

News, March 13, 2014. http://www.bbc.com/news/world-middle-east-26116868. Hamdan, Amani. “Women and Education in Saudi Arabia: Challenges and Achievements.”

International Education Journal 6, no. 1 (2005): 42–64. Hart, H. L. A. “Are There Any Natural Rights?” The Philosophical Review 64, no. 2 (April 1,

1955): 175–91. doi:10.2307/2182586. ———. The Concept of Law. London: Oxford University Press, 1961. Hart, Herbert Lionel Adolphus. “Positivism and the Separation of Law and Morals.” Harvard

Law Review, 1958, 593–629. Hobbes, Thomas. Leviathan: With Selected Variants from the Latin Edition of 1668. Edited by

Edwin Curley. Hackett Publishing Company, 1994. International Labor Organization. “Bangladesh Accord,” 2014. Accessed June 28, 2014,

http://www.bangladeshaccord.org/. ———. “ILO Declaration on Fundamental Principles and Rights at Work,” 1998.

http://www.ilo.org/declaration/thedeclaration/textdeclaration/lang--en/index.htm. James, Aaron. “A Practice Theory of Global Justice.” ms, 2014. ———. “Distributive Justice without Sovereign Rule : The Case of Trade.” Social Theory and

Practice 31, no. 4 (n.d.): 533–59. Accessed August 30, 2013. Kant, Immanuel, and Ted Humphrey. Perpetual Peace, and Other Essays on Politics, History,

and Morals. Indianapolis: Hackett Pub Co, 1983. Kelsen, Hans. Principles of International Law. Clark, N.J.: Lawbook Exchange, 2003. Kingsbury, Benedict, Nico Krisch, and Richard Stewart. The Emergence of Global

Administrative Law. SSRN Scholarly Paper. Rochester, NY: Social Science Research Network, April 14, 2005. http://papers.ssrn.com/abstract=692628.

Kolodny, Niko, and R. Jay Wallace. “Promises and Practices Revisited.” Philosophy & Public Affairs 31, no. 2 (April 1, 2003): 119–54.

Krasner, Stephen D. Sovereignty. Princeton University Press, 1999. ———. “Sovereignty An Institutional Perspective.” Comparative Political Studies 21, no. 1

(April 1, 1988): 66–94. Kratochwil, Friedrich. “Limits of Contract, The.” European Journal of International Law 5

(1994): 465. Kratochwil, Friedrich, Ed Mansfield, and John Gerard Ruggie. “International Organization: A

State of the Art on an Art of the State.” International Institutions: An International Organization Reader (Cambridge, MA: Massachusetts Institute of Technology and IO Foundation Press, 2001), 2001, 341–65.

140

Ku, Julian, and John Yoo. Taming Globalization: International Law, the U.S. Constitution, and the New World Order. Oxford University Press, USA, 2012.

Kumm, Mattias. “Constitutionalism and the Cosmopolitan State.” New York University Public Law and Legal Theory Working Papers, 2013. http://lsr.nellco.org/nyu_plltwp/423.

Lauterpacht, Hersch. The Function of Law in the International Community. Oxford University Press, 2011.

Locke, John. Second Treatise of Government. 1st ed. Hackett Pub Co, 1980. Mamdani, Mahmood. Good Muslim, Bad Muslim: America, the Cold War, and the Roots of

Terror. Reprint edition. New York: Harmony, 2005. Morgenthau, Hans J. Politics Among Nations, Revised by Kenneth W. Thompson. New York:

Knopf, 1985. Nagel, Thomas. “The Limits of International Law.” In Secular Philosophy and the Religious

Temperament. New York, NY: Oxford University Press, 2010. ———. “The Problem of Global Justice.” Philosophy & Public Affairs 33, no. 2 (2005): 113–47. Nozick, Robert. Anarchy, State, and Utopia. Second Edition edition. New York: Basic Books,

1975. Onuf, Nicholas. International Legal Theory: Essays and Engagements, 1966-2006. 1st ed.

Routledge-Cavendish, 2011. Organization for the Prohibition of Chemical Weapons. “Chemical Weapons Convention,” 1993.

Accessed June 28, 2014, http://www.opcw.org/chemical-weapons-convention/. Paul, Darel. “Sovereignty, Survival and the Westphalian Blind Alley in International Relations.”

Review of International Studies 25, no. 02 (1999): 217–31. Pogge, Thomas. “An Egalitarian Law of Peoples.” In Global Justice: Seminal Essays, edited by

Thomas Pogge and Darrel Moellendorf. St. Paul, MN: Paragon House, 2008. Prichard, Harold A., and J. A. Urmson. Moral Obligation and Duty and Interest: Essays and

Lectures. London, New York etc.: Oxford University Press, 1969. Rawls, John. A Theory of Justice: Original Edition. Belknap Press of Harvard University Press,

2005. ———. “Legal Obligation and the Duty of Fair Play.” In Law and Philosophy. New York

University Press, 1964. ———. “The Law of Peoples.” In Global Justice: Seminal Essays, edited by Thomas Pogge and

Darrel Moellendorf. St. Paul, MN: Paragon House, 2008. Raz, Joseph. “Authority, Law, and Morality.” The Monist 68, no. 3 (July 1, 1985): 295–324. ———. “Introduction.” In Authority, edited by Joseph Raz. New York, NY: New York

University Press, 1990. ———. “Problem of Authority: Revisiting the Service Conception, The.” Minn. L. Rev. 90

(2005): 1003. ———. The Authority of Law: Essays on Law and Morality. 2nd ed. Oxford University Press,

USA, 2009. Rocheleau, Jordy. “State Consent vs. Human Rights as Foundations for International Law.”

Social Philosophy Today 23 (July 1, 2007): 117–32. Rousseau, Jean-Jacques. The Social Contract. Translated by Maurice Cranston. Penguin

Classics, 1968. Ruggie, John Gerard. Constructing the World Polity: Essays on International

Institutionalisation. Routledge, 1998. S. S. Lotus (Fr. v. Turk.), PCIJ, 1927.

141

Scanlon, T. M. What We Owe to Each Other. Belknap Press of Harvard University Press, 2000. Shiffrin, Seana Valentine. “The Divergence of Contract and Promise.” Harvard Law Review 120,

no. 3 (January 1, 2007): 708–53. Simmons, A. John. Moral Principles and Political Obligations. Princeton, N.J.; Guildford:

Princeton University Press, 1981. Teslik, Lee Hudson. “NAFTA’s Economic Impact.” Council on Foreign Relations 7 (2009).

http://www.relooney.info/0_New_3568.pdf. United Nations. “Article 38.1 Statute of the International Court of Justice.” Accessed June 6,

2014. http://www.icj-cij.org/documents/index.php?p1=4&p2=2&p3=0&. Waltz, Kenneth Neal. Realism and International Politics. Routledge, 2008. Wendt, Alexander. “Anarchy Is What States Make of It: The Social Construction of Power

Politics.” International Organization 46, no. 02 (1992): 391–425. ———. Social Theory of International Politics. Cambridge University Press, 1999. Wendt, Alexander E. “The Agent-Structure Problem in International Relations Theory.”

International Organization 41, no. 03 (1987): 335–70. World Food Programme. “Hunger Statistics.” United Nations. Accessed June 6, 2014.

http://www.wfp.org/hunger/stats. Zemanek, Karl. “Vienna Convention on the Law of Treaties.” United Nations Audiovisual

Library of International Law, Accessed June 6, 2014, http://untreaty.un.org/cod/avl/pdf/ha/vclt/vclt-E. Pdf, 2009.